This is a prologue to reverting:
commit 3c66686
Author: Benjamin Walsh <benjamin.walsh@windriver.com>
Date: Tue Feb 9 17:34:02 2016 -0500
sys_clock: start the microkernel ticker in the MICROKERNEL init level
to allow the devices initializing in pre-MICROKERNEL init levels to poll
the hi-res clock (sys_cycle_get_32()), which relies on the system clock
having been started.
This change allows starting the system clock in the NANOKERNEL init
level by delaying announcing the ticks until the MICROKERNEL init level.
Change-Id: I43d54bb5e2f182d4edd880da0124a0817f911943
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Summary lines follow the @brief tag instead of the @file tag.
This prevents doxygen build warnings such as ...
The name xxx supplied as the second argument in the \file
statement is not an input file
Change-Id: I1014586ad21be12e14aa1d2a942e6b8a11211795
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
Modifications to timer drivers and interrupt setup, to manage
the tickless idle for the x86 architecture
Change-Id: Ie02d484b7e5636de6ea382ba2eeed57e704c8498
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
FIFOP setting is a wrong assumption to know if radio is on.
Transmission runs on clock basis instead of pre-set loop counter. Thus
removing useless config option relevantly as well.
Opmitizing a bit the code as well, and making cc2520_on() public for
testing purpose.
Change-Id: I4495d1d6c19d10dcbc18f7e2fd5041720ec1f438
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Integrating it into network stack. Centralizing all in drivers/802.15.4
Change-Id: Ia2916ff652afe5fe736f6560c2ed4a911a5f0679
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Adds extern "C" { } blocks to header files so that they can be
safely used by C++ source files.
Change-Id: Ia4db0c36a5dac5d3de351184a297d2af0df64532
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
The interrupt API has been redesigned:
- irq_connect() for dynamic interrupts renamed to irq_connect_dynamic().
It will be used in situations where the new static irq_connect()
won't work, i.e. the value of arguments can't be computed at build time
- a new API for static interrupts replaces irq_connect(). it is used
exactly the same way as its dynamic counterpart. The old static irq
macros will be removed
- Separate stub assembly files are no longer needed as the stubs are now
generated inline with irq_connect()
ReST documentation updated for the changed API. Some detail about the
IDT in ROM added, and an oblique reference to the internal-only
_irq_handler_set() API removed; we don't talk about internal APIs in
the official documentation.
Change-Id: I280519993da0e0fe671eb537a876f67de33d3cd4
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Rename class in pci_dev_info struct to allow to use C++ compilers.
Updated drivers to use new struct.
Change-Id: I17b94cb7bc094bccd615c8389a28589bfa90cab8
Signed-off-by: Javier B Perez Hernandez <javier.b.perez.hernandez@linux.intel.com>
The code to poke the system integration module to disable clock gating
for UARTs only works for UART0-3 since all the bits are in the same
register. However, clocks for UART4 and UART5 are controlled by
another register. This means that we have been writing to the wrong
bit for enabling UART4.
This patch fixes this issue, and moves the clock gating clock into
board initialization. The incorrect code has also been removed to
prevent accidental mis-use. The dev_data struct is no longer needed
for uart_k20, so that is removed as well.
Change-Id: I67845a417e43647bf0ffcbdbda34ce68fa887713
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Add support for compilers conforming to the IAMCU calling convention
as documented by
https://github.com/hjl-tools/x86-psABI/wiki/iamcu-psABI-0.7.pdf
Change-Id: I6fd9d5bede0538b2049772e3850a5940c5dd911e
Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
All of these stubs at the end pop the stack and tell the LOAPIC
that we are at EOI. Put this in common code which falls through to
_IntExit(). Saves 8 bytes per static IRQ stub. There is also small
constant savings in the common code for dynamic IRQs.
Change-Id: If17e9f105928a4251a2cb3fc0d192649c1c4d84b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Quark SE Lakemont core has a hardware bug where the LOAPIC does
not properly notify the IOAPIC to clear the IRR bit for level-
triggered interrupts.
This patch introduces a workaround where the vector ID of the
in-service interrupt is manually written to the IOAPIC_EOI
register, resulting in the bit being cleared.
Unfortunately, in the context where EOI happens it's very difficult
to identify which IRQ line is being serviced, so this is done
unconditionally for all interrupts vectors whether they are registered
in the IOAPIC RTE table or not.
Change-Id: I639cd258dec4f50934e17eadbb821e6a7112e636
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This was only needed for an older implementation of software interrupts,
now superseded by the irq_offload() API (which doesn't interact with
the interrupt controller at all)
Change-Id: I8aa696d370ae1799872f6d70de69f3cb5b47456a
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
To many people, IPI connotes inter-processor interrupts on SMP
systems. Rename this to IPM, or Inter-Processor Mailboxes.
Change-Id: I032815e23c69a8297c0a43992132441c240fb71e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Other internal functions are shown in this header, no reason to
keep this a secret.
Change-Id: Icb7d36206148c281f1960d1ac10368d9bb3033f1
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The initialization code and macros are the same. So extracts it
and puts it into the driver. This is another step to follow
the driver model.
Those empty C files are there because the current Kbuild
requires Makefile to be present at those directories,
(due to arch/x86/Kbuild), which requires building some object
files. So the empty source files are there to produce empty
object files to satisfy this.
Change-Id: I14056347ea14cff227d9e8960192e8673c0019b8
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The initialization code and macros are the same. So extracts it
and puts it into the driver. This is another step to follow
the driver model.
Change-Id: I1d379068f64855d5d4595838040ec50f97f638a0
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The initialization code and macros are the same. So extracts it
and puts it into the driver. This is another step to follow
the driver model.
Change-Id: I1af8b2888779b2b58367feaff9ee1a6d97b4873c
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
It's very rare to have lines exceeding 100 characters (default console
width being 80) so 1024 is quite overkill. Even 256 is probably more
than anyone will need any time soon, but if that becomes an issue this
define can be converted to a Kconfig option in the future.
Change-Id: If5fd324b17a0e8ef7b39ea7cc6624739d0241cac
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Removed old style file description and documnetation and apply
doxygen synatx.
Change-Id: I3ac9f06d4f574bf3c79c6f6044cec3a7e2f6e4c8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The content of include/drivers/hpet.h is really private to
the driver and does not need to be exposed in public include
directory. So move the content into driver code.
Change-Id: Ica442e43c480a6b079b8d3c4e75e67adcfd0ba6b
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
() Rename nsim_uart.c to uart_nsim.c. This is to follow
the driver naming convention.
() Rename functions nsim_uart_*() to uart_nsim_*(),
following driver naming convention.
() UART ports initialization is moved into the driver itself.
All the init code in platform config files is removed.
() Adds (many) Kconfig options. These don't have to be defined
in each platform's board.h anymore.
Change-Id: If015f39a6f6b4fcc65625e6e5f973b4469202f54
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
() Rename k20UartDrv.c to uart_k20.c. This is to follow the driver
naming convention.
() Move driver/serial/k20_uart.h to drivers/serial/uart_k20_priv.h
as this contains definitions private to the driver, and
should not be exposed in public include directory.
() Rename functions k20_uart_*() to uart_k20_*(), following
driver naming convention.
() Renames CONFIG_K20_UART_* to CONFIG_UART_K20_*
() UART ports initialization is moved into the driver itself.
All the init code in platform config files is removed.
() Adds (many) Kconfig options. These don't have to be defined
in each platform's board.h anymore.
Change-Id: If1be1fde083aba6ff68062db2059aef08617a286
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Renames _sys_clock_cycle_get() (provided by the timer driver)
to sys_cycle_get_32(). It is the preferred method to read the
hardware clock.
Change-Id: Ifea5213d8c04a8bf7b9114b048c5db0ccee61549
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
Replaces microkernel global variable _k_sys_clock_tick_count with the
global variable _sys_clock_tick_count. This allows both the microkernel
and the nanokernel to use the same variable to track system clock ticks
instead of using two different ones.
Change-Id: Ia4eebf022f59d130ad1882e0e550016527543a45
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
Fixes the type of global variable _sys_idle_elapsed_ticks such that
both its nanokernel and microkernel definitions are of the same
type--int32_t.
One of the repercussions of this is that code related to the routine
_nano_sys_clock_tick_announce() does not need the application of
typecasting.
Change-Id: I3d7374cd1a32aea7e4651726febde74ebe4ac8ac
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
Improves documentation consistency by using "N/A" instead of "None"
when the routine does not return a value.
Change-Id: I429a159b3037742cbc431db0cb4828ab9d6d35c3
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
Original name was too generic and confusing. This patch renames
driver to pipe UART and moves it to console drivers folder. Kconfig
destription is also improved.
Change-Id: I716fdbf7d636bbdc03b0fce27a59fd866f473246
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
This module is based on the standard Local APIC and IO APIC source modules.
This modules combines these modules into one source module that exports the
same APIs defined by the Local APIC and IO APIC header modules. These
routine have been adapted for the Quark D2000 Interrupt Controller which has
a cutdown implementation of the Local APIC & IO APIC register sets.
Change-Id: Ic80aa78918483663d76054ebadefa08d8a3f188a
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Quark SE provides various clock controllers through its SCSS block.
Peripheral, external, sensor, and others.
This current drivers provides only the clock gating capability, for
peripheral, external and sensor. But it could support divider and more
other features once defined in the generic API.
Note: such clock has _nothing_ to do with a Real Time Clock (RTC).
An RTC provide clock timing like a watch would do. Here the clock
controller is about circuit clocking.
Change-Id: I1a365ae730dfc6be7686271f7fbb693e64a6ff6f
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
__unused is define in sys/cdefs.h when building against newlib.
Change-Id: I0fad81dc91e1eb91368b5b3370e3fbb0fb7ba4a9
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Rename them to SYSTEM_CLOCK_DISABLE and sys_clock_disable to put
the symbols in the proper namespace and to align with the rest of
the timer APIs.
Change-Id: I341cdb1bd6cc01e5662d8526c293d4e0acd6ec24
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
We generally call this in interrupt code, and since it's just a register
write, get rid of all the function call overhead.
Change-Id: I251fa3cee0841fcdd8eecc7941d4d3782d112e66
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
We are interested in supporting some XIP x86 platforms which are
unable to fetch CPU instructions from system RAM. This requires
refactoring our dynamic IRQ/exc code which currently synthesizes
assembly language instructions to create IRQ stubs on-the-fly.
Instead, a new approach is taken. Given that the configuration at
build time specifies the number of required stubs, use this
to generate a build time a set of tiny stub functions which simply
push a 'stub id' and then call common dynamic interrupt code.
The handler function and handler argument is saved in a table keyed by
this stub id.
CONFIG_EOI_HANDLER_SUPPORTED removed, the code hasn't been conditionally
compiled for some time and in all cases we call _loapic_eoi() when
finished with an interrupt.
Some other out-of-date verbiage in comments related to supporting
non-APIC removed.
Previously, when dynamic exceptions were created a pointer would
be passed in by the caller reserving ram for the stub code. Since
this is no longer feasible, two new Kconfig options have been added.
CONFIG_NUM_DYNAMIC_EXC_STUBS and CONFIG_NUM_DYNAMIC_EXC_NO_ERR_STUBS
control how many stubs are created for exceptions that push
an error code, and no error code, respectively.
SW Interrupts are no longer triggered by "int <vector>" hard-coded
assembly instructions. Instead this is done by sending a self-directed
inter-processor interrupt from the LOAPIC, using a new API
loapic_int_vect_trigger(). In this way we get rid of dynamically
generated code in irq_test_common.h.
All interrupts call _loapic_eoi() when finished, since this is now
the right thing to do for all IRQs, including SW interrupts.
_irq_handler_set() for x86 no longer requires the old function pointer
to be supplied.
Change-Id: I78993d3d00dd153c9051c518b417cce8d3acee9e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The tick event is now generated by calling the standard API for
signalling an event from an ISR. (This allows the tick event
to be generated properly even if the internal implementation of
the event sub-system changes.)
Change-Id: Ia2be3874358acb0ea2b2d2bbefbe4603fe11da8a
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Removing many of the typedefs that are only used once to lessen the
checkpatch warning about creating new typedefs. A handful have been
behind as they would require a more invasive change to the code. It
has yet to be determined if this is a worthwhile endavour.
Change-Id: Ibeb29e0a1d37e8121218fccf0d986cbebd226e85
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
This patch adds the pci_enable_bus_master function that enables bus
master access for a PCI device.
Change-Id: I317daac1a613ccfc5a3894332c1a7dbbff05991f
Signed-off-by: Michael LeMay <michael.lemay@intel.com>
Includes device.h to prevent potential compiler warnings due to use of
'struct device *' in _ioapic_init() declaration.
Change-Id: Iabd97c09385044a3742658f66a79be969b62852a
Work-by: Johan Kruger <johan.kruger@windriver.com>
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
This was only needed for the older 8259A style PICs which are no
longer supported.
Since we now just support APIC, we always just call loapic_eoi which
no longer requires an argument and informs the IOAPIC that the interrupt
is complete if necessary.
Change-Id: I15c9b7b4f03b872656220af32220b62e043bfa6b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Change-Id: I2276676142deea21cf8079449ce153f2fb887a8e
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Change-Id: I819d13f0d7a23e3a61dcda6a3ced18810b192158
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Change-Id: I6da43e41f9c6efee577b70513ec368ae3cce0144
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Only driver specific public headers should be found in include/drivers.
All generic API are found in include/ directory.
Change-Id: Ic50931987bb9460fd4a3843abc6f5de107faf045
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Add a context value to the ioapic and loapic mkstub macros to allow
the developer to specify a context value that will be passed on the
stack to the interrupt service routines.
All the invocations for the loapic_mkstub and ioapic_mkstub macros
have been changed to pass in 0 so there is no functional difference
ATM. This change removes the need for drivers to have trampoline ISR
routines to pass a context vaule to their generic ISR routine.
This is the first step in getting rid of the hack where the driver
needs to look into the __initconfig_* linker objects.
Change-Id: I2c5eaa20d8cb5a42ef445762c426854be32c8452
Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
Change all the Intel and Wind River code license from BSD-3 to Apache 2.
Change-Id: Id8be2c1c161a06ea8a0b9f38e17660e11dbb384b
Signed-off-by: Javier B Perez Hernandez <javier.b.perez.hernandez@linux.intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
