Most kernel APIs are now ready for inclusion in the API guide.
The APIs largely follow a standard template to provide users
of the API guide with a consistent look-and-feel.
Change-Id: Ib682c31f912e19f5f6d8545d74c5f675b1741058
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Updates x86 floating point support to reflect changes that have
been made in recent months.
* Many, many, many cosmetic changes (mostly revisions to comments).
* Elimination of unnecessary function aliases that were needed
to support the task and fiber versions of certain APIs.
* Elimination of run-time code to enable a thread's "FP regs"
option bit if the "SSE regs" option bit was set. The kernel
now recognizes that the thread is using the FPU as long as
either option bit is set. (If the thread has both option bits
enabled this is the same as if only the "SSE regs" bit is set.)
Change-Id: Ic12abc54b6fa78921749b546d8debf23e7ad232d
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
C++ support moved from nanokernel.h to kernel.h.
Change-Id: I5e1631941e26f4ab3f311b680267b743bab15e40
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Symbols now use the K_ prefix which is now standard for the
unified kernel. Legacy support for these symbols is retained
to allow existing applications to build successfully.
Change-Id: I3ff12c96f729b535eecc940502892cbaa52526b6
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
There are a number of data sections that are repeated across
all the linker scripts for various architecture. In practice these
don't always get updated and we have had problems with bit-rot.
Consolidate these to make maintenance easier.
x86 linker scripts now follow the same naming convention and we
get rid of a linker-epilog.h that wasn't necessary and whose purpose
has been lost to the mists of time. If applications want to define their
own sections they should be allowed to. Linker scripts for x86 do not
end with .h any more, they are not C header files even though we use
C's preprocessor.
Issue: ZEP-688
Change-Id: I893eb4619969695c1f980efd7c2ec9fa5dad136d
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
This is used by a test case, and it's better to just put this
here instead of forking the linker scripts.
Change-Id: Ifbb90b73bb26118ae2422cc6feccb3db58a26f2c
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
This mechanism was intended to reserve space during the first pass for
certain data structures created by gen_idt, but this is unnecessary.
The only memory addresses that must be fixed between the two passes are the
locations of the interrupt stubs, which are in the .text section much
earlier than the generated data structures; they do not shift.
Change-Id: I3aab00e171e6a9ff439a7af8d69769e4c29337a7
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Binutils ld has an annoying misfeature (apparently a regression from a
few years ago) that alignment directives (and alignment specifiers on
symbols) apply only to the runtime addresses and not, apparently, to
the load address region specified with the "AT>" syntax. The net
result is that by default the LMA output ends up too small for the
addresses generated in RAM. See here for some details:
https://sourceware.org/ml/binutils/2013-06/msg00246.htmlhttps://sourceware.org/ml/binutils/2014-01/msg00350.html
The required workaround/fix is that AFAICT any section which can have
inherit a separate VMA vs. LMA from a previous section must specify an
"ALIGN_WITH_INPUT" attribute. Otherwise the sections will get out of
sync and the XIP data will be wrong at runtime.
No, I don't know why this isn't the default behavior.
A further complexity is that this feature only works as advertised
when the section is declared with the "AT> region" syntax after the
block and not "AT(address)" in the header. If you use the header
syntax (with or without ALIGN_WITH_INPUT), ld appears to DOUBLE-apply
padding and the LMA ends up to big. This is almost certainly a
binutils bug, but it's trivial to work around (and the working syntax
is actually cleaner) so we adjust the usage here.
Note finally that this patch includes an effective reversion of commit
d82e9dd9 ("x86: HACK force alignment for _k_task_list section"), which
was an earlier workaround for what seems to be the same issue.
Jira: ZEP-955
Change-Id: I2accd92901cb61fb546658b87d6752c1cd14de3a
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Interrupt stubs now just push the ISR and parameter onto the stack
and jump to the common interrupt code, never to return.
Change-Id: I82543d8148b5c7dfe116c43f41791f852614bb28
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Due to the memory pool structure only static declaration of
memory pool is possible.
Change-Id: I4797ed88fd2ac3b7812ff26e552e1745611c4575
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
This header has a bunch of data structure definitions and macros useful
for manipulating segment descriptors on X86. The old IDT_ENTRY defintion
is removed in favor of the new 'struct segment_descriptor' which can be
used for all segment descriptor types and not just IRQ gates.
We also add some inline helper functions for examining segment registers,
descriptor tables, and doing far jumps/calls.
Change-Id: I640879073afa9765d2a214c3fb3c3305fef94b5e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The Arduino 101 comes with a bootloader that supports DFU
and flashing of all cores using the dfu-util package.
This changes the memory layout of the image built for the
Arduino 101 and remove previous work-arounds to allow booting,
including the version-header section in the linker script.
The bootloader expects the text section at +0x30 from the physical
load address and thus requires special treatment in the linker
script.
Other changes by Andrew Boie:
The flash size parameters were both wrong. X86 side has 192K
of flash from 0x4003000 - 0x40060000, the entire span of
sys_flash1.
ARC side is now the span from 0x40010000 - 0x40030000, 128K.
Change-Id: Iecfa5d2b84a3f522d9eca06268d6b8b71a094aaa
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
In 1.0 you could set only one callback on the whole gpio controller. It
was impossible for another sub-system to add another callback, without
overwritting an existing one.
Such API has been obsolete for a long time and no one is using it
anymore. Thus removing it entirely.
Change-Id: I6a17fd99373dc6cef1fa2ebb421e992412d5015e
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
The x86 architecture port is fitted with support for the unified kernel,
namely:
- the interrupt exit code now calls _Swap() if the current
thread is not a coop thread and if the scheduler is not locked
- there is no 'task' fields in the _nanokernel anymore: _Swap()
now calls _get_next_ready_thread instead
- the _nanokernel.fiber field is replaced by a more sophisticated
ready_q, based on the microkernel's priority-bitmap-based one
- nano_private includes nano_internal.h from the unified directory
- the FIBER, TASK and PREEMPTIBLE flags do not exist anymore: the thread
priority drives the behaviour
- the tcs uses a dlist for queuing in both ready and wait queues instead
of a custom singly-linked list
- other new fields in the tcs include a schedule-lock count, a
back-pointer to init data (when the task is static) and a pointer to
swap data, needed when a thread pending on _Swap() must be passed more
then just one value (e.g. k_stack_pop() needs an error code and data)
- fiberRtnValueSet() is aliased to _set_thread_return_value since it
also operates on preempt threads now
- _set_thread_return_value_with_data() sets the swap_data field in
addition to a return value from _Swap()
- convenience aliases are created for shorter names:
- _current is defined as _nanokernel.current
- _ready_q is defined as _nanokernel.ready_q
- _Swap() sets the threads's return code to -EAGAIN before swapping out
to prevent timeouts to have to set it (solves hard issues in some
kernel objects).
- Floating point support.
Note that, in _Swap(), the register holding the thread to be swapped in has
been changed from %ecx to %eax in both the legacy kernel and the unified kernel
to take advantage of the fact that the return value of _get_next_ready_thread()
is stored in %eax, and this avoids moving it to %ecx.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: I4ce2bd47bcdc62034c669b5e889fc0f29480c43b
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Commit 3e63a74514 did not revert properly
things.
Change-Id: I792d5698966542ce2cfb9f858c56b30c392f02a2
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.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>
Originally, x86 just supported APIC. Then later support
for the Mint Valley Interrupt Controller was added. This
controller is mostly similar to the APIC with some differences,
but was integrated in a somewhat hacked-up fashion.
Now we define irq_controller.h, which is a layer of abstraction
between the core arch code and the interrupt controller
implementation.
Contents of the API:
- Controllers with a fixed irq-to-vector mapping define
_IRQ_CONTROLLER_VECTOR_MAPPING(irq) to obtain a compile-time
map between the two.
- _irq_controller_program() notifies the interrupt controller
what vector will be used for a particular IRQ along with triggering
flags
- _irq_controller_isr_vector_get() reports the vector number of
the IRQ currently being serviced
- In assembly language domain, _irq_controller_eoi implements
EOI handling.
- Since triggering options can vary, some common defines for
triggering IRQ_TRIGGER_EDGE, IRQ_TRIGGER_LEVEL, IRQ_POLARITY_HIGH,
IRQ_POLARITY_LOW introduced.
Specific changes made:
- New Kconfig X86_FIXED_IRQ_MAPPING for those interrupt controllers
that have a fixed relationship between IRQ lines and IDT vectors.
- MVIC driver rewritten per the HAS instead of the tortuous methods
used to get it to behave like LOAPIC. We are no longer writing values
to reserved registers. Additional assertions added.
- Some cleanup in the loapic_timer driver to make the MVIC differences
clearer.
- Unused APIs removed, or folded into calling code when used just once.
- MVIC doesn't bother to write a -1 to the intList priority field since
it gets ignored anyway
Issue: ZEP-48
Change-Id: I071a477ea68c36e00c3d0653ce74b3583454154d
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
This reverts commit d73a9bb9c6.
The patch was intended for 1.6.0 release.
Change-Id: Id42058b746a3d2a54e4b1a2983eb58bd10b1ed40
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Previously, exception stubs had to be declared in assembly
language files. Now we have two new APIs to regsiter exception
handlers at C toplevel:
_EXCEPTION_CONNECT_CODE(handler, vector)
_EXCEPTION_CONNECT_NOCODE(handler, vector)
For x86 exceptions that do and do not push error codes onto
the stack respectively.
In addition, it's now no longer necessary to #define around
exception registration. We now use .gnu.linkonce magic such that
the first _EXCEPTION_CONNECT_*() that the linker finds is used
for the specified vector. Applications are free to install their
own exception handlers which will take precedence over default
handlers such as installed by arch/x86/core/fatal.c
Some Makefiles have been adjusted so that the default exception
handlers in arch/x86/core/fatal.c are linked last. The code has
been tested that the right order of precedence is taken for
exceptions overridden in the floating point, gdb debug, or
application code. The asm SYS_NANO_CPU_EXC_CONNECT API has been
removed; it was ill- conceived as it only worked for exceptions
that didn't push error codes. All the asm NANO_CPU_EXC_CONNECT_*
APIs are gone as well in favor of the new _EXCEPTION_CONNNECT_*()
APIs.
CONFIG_EXCEPTION_DEBUG no longer needs to be disabled for test
cases that define their own exception handlers.
Issue: ZEP-203
Change-Id: I782e0143fba832d18cdf4daaa7e47820595fe041
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Rather than embedding the ISR stub directly inside the function that
invokes IRQ_CONNECT(), stick all the generated stubs in the
.text.irqstubs section.
In this way, we make things easier to debug since the stub code isn't
mixed in with the "calling" function's assembly, and we no longer
need an instruction to jump over it.
Since these are now in their own section and not embedded inside an
unrelated init function, we unconditionally generate descriptive
symbol names for each stub based on the name of the handler and the
IRQ line.
Example for HPET timer on IRQ #2:
00100440 T _timer_int_handler <-- driver ISR
00100590 T _timer_int_handler_irq2_stub <-- generated stub
Change-Id: I49425aef7775edbca8ad7f61d2d4f9c41cb0d39d
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Follow up to TSC decission for further discussion in the networking
WIG.
Change-Id: I148b484dfe308661573e47ed3e60cceed673bddf
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Having two parallel implementations is a maintenance issue, especially
when some strategically placed #ifdefs will suffice.
We prefer the ASM versions for SYS V, as we need complete control of
the emitted assembly for interrupt handling and context switching.
The SYS V code is far more mature. IAMCU C code has known issues with
-fomit-frame-pointer.
The only difference between the two calling conventions is that the
first three function arguments are provided in eax, edx, ecx instead
of on the stack.
Issue: ZEP-49
Change-Id: I9245e4b0ffbeb6d890a4f08bc8a3a49faa6d8e7b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Net core then does not know anything about l2 related logic.
For instance ARP is used in ethernet l2 API and nowhere else.
This will be helpful when adding different technologies altogether.
Currently, only SLIP driver is enabled to use relevant l2 layer.
Change-Id: I03c93326321028d04222733ca4083e3c6b785202
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
This will be used by the new network stack to relate a device to actual
network context, and used in the different layers (mac, ip ...).
Change-Id: I30c08fa975314544c36b71636fd9653d562891b3
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
The problem is doxygen's parser is getting confused by constructs as:
static inline __attribute__((always_inline))
void sys_out8(uint8_t data, io_port_t port)
{
_arc_v2_aux_reg_write(port, data);
}
Too many words at the beginning of the function definition. So change
to use the macro ALWAYS_INLINE (which is already defined to mean
'inline __attribute__((always_inline))`.
Kills:
sys_io.h:37: warning: documented symbol `static inline void sys_out8' was not declared or defined.
sys_io.h:47: warning: documented symbol `static inline uint8_t sys_in8' was not declared or defined.
sys_io.h:58: warning: documented symbol `static inline void sys_out16' was not declared or defined.
sys_io.h:68: warning: documented symbol `static inline uint16_t sys_in16' was not declared or defined.
sys_io.h:79: warning: documented symbol `static inline void sys_out32' was not declared or defined.
sys_io.h:89: warning: documented symbol `static inline uint32_t sys_in32' was not declared or defined.
sys_io.h:120: warning: documented symbol `static inline int sys_io_test_bit' was not declared or defined.
sys_io.h:133: warning: documented symbol `static inline int sys_io_test_and_set_bit' was not declared or defined.
sys_io.h:146: warning: documented symbol `static inline int sys_io_test_and_clear_bit' was not declared or defined.
sys_io.h:161: warning: documented symbol `static inline void sys_write8' was not declared or defined.
sys_io.h:171: warning: documented symbol `static inline uint8_t sys_read8' was not declared or defined.
sys_io.h:182: warning: documented symbol `static inline void sys_write16' was not declared or defined.
sys_io.h:192: warning: documented symbol `static inline uint16_t sys_read16' was not declared or defined.
sys_io.h:248: warning: documented symbol `static inline int sys_test_bit' was not declared or defined.
sys_io.h:261: warning: documented symbol `static inline int sys_test_and_set_bit' was not declared or defined.
sys_io.h:274: warning: documented symbol `static inline int sys_test_and_clear_bit' was not declared or defined.
Change-Id: Id10e9b6cd44a370ccc732c17b23fb66bd1845205
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
In scenarios where device PM is enabled and dynamic irqs are
used, move the irq to vector table to RAM and keep it updated,
so that we can use this to restore IOAPIC/LOAPIC vector entries.
Jira: ZEP-224
Change-Id: I0d4350d4e30f8ca337a2a1d4f012748c3cb450f4
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
The stub label is created with ISR and IRQ number since the same
ISR can be used by several IRQs
Change-Id: I0ea909fddbce7a70c754befd095b7a3b36fffab4
Signed-off-by: Fabrice Olivero <fabrice.olivero@intel.com>
To boot zephyr on the Arduino 101 running the original bootloader
which supports DFU, set the following in your application configuration
file:
CONFIG_SS_RESET_VECTOR=0x40034000
CONFIG_PHYS_LOAD_ADDR=0x40010000
CONFIG_VERSION_HEADER=y
Jira: ZEP-219
Change-Id: Ia015a7b6fce888b49ed22c558de992132d4713ea
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Many sub-systems might require to set a callback on different pins.
Thus enabling it via changing the API.
It is also possible to retrieve private-data in the callback handler
using CONTAINER_OF() macro (include/misc/util.h).
Former API is still available, and is emulated through the new one.
Using both should not be a problem as it's using new API calls.
However, it's now better to start using the new API.
Change-Id: Id16594202905976cc524775d1cd3592b54a84514
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Introduce an x86 interrupt stack frame that contains more information
than the non-debug one, namely the caller-saved GPRs, as well as an API
to retrieve it. Able to handle nested interrupts stack frames.
Change-Id: If182aaa2f34e4714b16ca65ff79da63b72d962f7
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
It wasn't correct to add the size of the long jump instruction
as it *replaces* a short jump instead of just being after it.
So redefine this to be the difference in size between these
two instructions.
Change-Id: I65be2afab19d9cd8b096551acde0156f0503df87
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Use of these is the mark of a deranged imagination.
Change-Id: Ib4b5f78cf61c016e333288090b397e9a3e0b8a40
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
These are guaranteed to work for bitfields that are
larger then 32 bits wide.
Change-Id: I39a641f08a255478fae583947bced762950d12ff
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The app-facing interface for configuring interrupts was never
formally defined, instead it was defined separately for each arch
in their respective arch-specific header files. Occasionally these
would go out of sync.
Now there is a single irq.h header which defines this interface.
To avoid runtime overhead, these map to _arch_* implementations of
each that must be defined in headers pulled in by arch/cpu.h.
Change-Id: I69afbeff31fd07f981b5b291f3c427296b00a4ef
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Looking at the IDT in a debugger is confusing, add a pretty-printing
function.
Change-Id: Iacc5e204e5d11e3e875c75ddf6d2e2e80b230299
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Having priority levels 0 and 1 reserved on x86 due to implementation
details on how the CPU uses the vector table is confusing to users,
and makes it unnecessarily difficult to share drivers between arches.
Now on x86, priority levels 0 and 1 are available. Semantically, all
priority levels have had 2 subtracted from them.
It is no longer necessary to specify a priority level when the
vector itself is specified. If an IDT entry has a specific vector
associated with it, any priority argument is simply ignored.
In gen_idt, some simplifications have been made:
- The printed representation of a generated entry now fits on one line
- Some checks being done in validate_priority() were redundant, as
generate_interrupt_vector_bitmap() also ensures that there are
sufficient free vectors within a priority level.
Change-Id: I26669d8ee0a53f48fbc2283490a8c42d8b1daf8e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
It's not a function and requires all its arguments to be build-time
constants. Make this more obvious to the end user to ease confusion.
Change-Id: I64107cf4d9db9f0e853026ce78e477060570fe6f
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Most systems have far less than 256 IRQ lines available, so
save some bytes in ROM by making this a config option.
On systems with MVIC, omit the table entirely as the mapping
is fixed.
The build cmd_gen_idt is slightly easier to read and will fail
immediately if any of the commands in the sequence error out.
Change-Id: I411f114557591e5cd96b618e6f79f97e8bedadf0
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
This bitfield is only needed to find unused vectors in the IDT
for installing dynamic interrupts.
Change-Id: I34ecd330774a0e50f240b4396527682eded29627
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Allan Stephens