unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
/*
|
|
|
|
* Copyright (c) 2016 Wind River Systems, Inc.
|
|
|
|
*
|
|
|
|
* Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
* you may not use this file except in compliance with the License.
|
|
|
|
* You may obtain a copy of the License at
|
|
|
|
*
|
|
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
*
|
|
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
|
|
* distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
* See the License for the specific language governing permissions and
|
|
|
|
* limitations under the License.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <kernel.h>
|
2016-11-08 10:36:50 -05:00
|
|
|
#include <kernel_structs.h>
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
#include <atomic.h>
|
2016-10-13 10:31:48 -04:00
|
|
|
#include <ksched.h>
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
#include <wait_q.h>
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
/* the only struct _kernel instance */
|
|
|
|
struct _kernel _kernel = {0};
|
|
|
|
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
/* set the bit corresponding to prio in ready q bitmap */
|
|
|
|
static void _set_ready_q_prio_bit(int prio)
|
|
|
|
{
|
|
|
|
int bmap_index = _get_ready_q_prio_bmap_index(prio);
|
2016-11-08 10:36:50 -05:00
|
|
|
uint32_t *bmap = &_ready_q.prio_bmap[bmap_index];
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
*bmap |= _get_ready_q_prio_bit(prio);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* clear the bit corresponding to prio in ready q bitmap */
|
|
|
|
static void _clear_ready_q_prio_bit(int prio)
|
|
|
|
{
|
|
|
|
int bmap_index = _get_ready_q_prio_bmap_index(prio);
|
2016-11-08 10:36:50 -05:00
|
|
|
uint32_t *bmap = &_ready_q.prio_bmap[bmap_index];
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
*bmap &= ~_get_ready_q_prio_bit(prio);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Add thread to the ready queue, in the slot for its priority; the thread
|
|
|
|
* must not be on a wait queue.
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
*
|
|
|
|
* This function, along with _move_thread_to_end_of_prio_q(), are the _only_
|
|
|
|
* places where a thread is put on the ready queue.
|
|
|
|
*
|
|
|
|
* Interrupts must be locked when calling this function.
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
*/
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-10-05 17:32:01 -04:00
|
|
|
void _add_thread_to_ready_q(struct k_thread *thread)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
int q_index = _get_ready_q_q_index(thread->base.prio);
|
|
|
|
sys_dlist_t *q = &_ready_q.q[q_index];
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
_set_ready_q_prio_bit(thread->base.prio);
|
|
|
|
sys_dlist_append(q, &thread->base.k_q_node);
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
struct k_thread **cache = &_ready_q.cache;
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
*cache = *cache && _is_prio_higher(thread->base.prio,
|
|
|
|
(*cache)->base.prio) ?
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
thread : *cache;
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
|
|
|
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
/*
|
|
|
|
* This function, along with _move_thread_to_end_of_prio_q(), are the _only_
|
|
|
|
* places where a thread is taken off the ready queue.
|
|
|
|
*
|
|
|
|
* Interrupts must be locked when calling this function.
|
|
|
|
*/
|
|
|
|
|
2016-10-05 17:32:01 -04:00
|
|
|
void _remove_thread_from_ready_q(struct k_thread *thread)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
int q_index = _get_ready_q_q_index(thread->base.prio);
|
|
|
|
sys_dlist_t *q = &_ready_q.q[q_index];
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
sys_dlist_remove(&thread->base.k_q_node);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
if (sys_dlist_is_empty(q)) {
|
2016-11-08 10:36:50 -05:00
|
|
|
_clear_ready_q_prio_bit(thread->base.prio);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
struct k_thread **cache = &_ready_q.cache;
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
|
|
|
*cache = *cache == thread ? NULL : *cache;
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
/* reschedule threads if the scheduler is not locked */
|
|
|
|
/* not callable from ISR */
|
|
|
|
/* must be called with interrupts locked */
|
|
|
|
void _reschedule_threads(int key)
|
|
|
|
{
|
|
|
|
K_DEBUG("rescheduling threads\n");
|
|
|
|
|
|
|
|
if (_must_switch_threads()) {
|
|
|
|
K_DEBUG("context-switching out %p\n", _current);
|
|
|
|
_Swap(key);
|
|
|
|
} else {
|
|
|
|
irq_unlock(key);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-11-10 14:46:58 -05:00
|
|
|
void k_sched_lock(void)
|
|
|
|
{
|
|
|
|
__ASSERT(!_is_in_isr(), "");
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
atomic_inc(&_current->base.sched_locked);
|
2016-11-10 14:46:58 -05:00
|
|
|
|
|
|
|
K_DEBUG("scheduler locked (%p:%d)\n",
|
2016-11-08 10:36:50 -05:00
|
|
|
_current, _current->base.sched_locked);
|
2016-11-10 14:46:58 -05:00
|
|
|
}
|
|
|
|
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
void k_sched_unlock(void)
|
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
__ASSERT(_current->base.sched_locked > 0, "");
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
__ASSERT(!_is_in_isr(), "");
|
|
|
|
|
|
|
|
int key = irq_lock();
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
atomic_dec(&_current->base.sched_locked);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
K_DEBUG("scheduler unlocked (%p:%d)\n",
|
2016-11-18 16:08:24 -05:00
|
|
|
_current, _current->base.sched_locked);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
_reschedule_threads(key);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Callback for sys_dlist_insert_at() to find the correct insert point in a
|
|
|
|
* wait queue (priority-based).
|
|
|
|
*/
|
2016-11-08 10:36:50 -05:00
|
|
|
static int _is_wait_q_insert_point(sys_dnode_t *node, void *insert_prio)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
2016-10-05 17:32:01 -04:00
|
|
|
struct k_thread *waitq_node =
|
2016-11-08 10:36:50 -05:00
|
|
|
CONTAINER_OF(
|
|
|
|
CONTAINER_OF(node, struct _thread_base, k_q_node),
|
|
|
|
struct k_thread,
|
|
|
|
base);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
return _is_prio_higher((int)insert_prio, waitq_node->base.prio);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
/* convert milliseconds to ticks */
|
|
|
|
|
|
|
|
#define ceiling(numerator, divider) \
|
|
|
|
(((numerator) + ((divider) - 1)) / (divider))
|
|
|
|
|
|
|
|
int32_t _ms_to_ticks(int32_t ms)
|
|
|
|
{
|
|
|
|
int64_t ms_ticks_per_sec = (int64_t)ms * sys_clock_ticks_per_sec;
|
|
|
|
|
|
|
|
return (int32_t)ceiling(ms_ticks_per_sec, MSEC_PER_SEC);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* pend the specified thread: it must *not* be in the ready queue */
|
|
|
|
/* must be called with interrupts locked */
|
2016-10-05 17:32:01 -04:00
|
|
|
void _pend_thread(struct k_thread *thread, _wait_q_t *wait_q, int32_t timeout)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
|
|
|
sys_dlist_t *dlist = (sys_dlist_t *)wait_q;
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
sys_dlist_insert_at(dlist, &thread->base.k_q_node,
|
|
|
|
_is_wait_q_insert_point,
|
|
|
|
(void *)thread->base.prio);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
_mark_thread_as_pending(thread);
|
|
|
|
|
|
|
|
if (timeout != K_FOREVER) {
|
|
|
|
_mark_thread_as_timing(thread);
|
2016-10-17 14:34:53 -05:00
|
|
|
_add_thread_timeout(thread, wait_q,
|
|
|
|
_TICK_ALIGN + _ms_to_ticks(timeout));
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* pend the current thread */
|
|
|
|
/* must be called with interrupts locked */
|
|
|
|
void _pend_current_thread(_wait_q_t *wait_q, int32_t timeout)
|
|
|
|
{
|
|
|
|
_remove_thread_from_ready_q(_current);
|
|
|
|
_pend_thread(_current, wait_q, timeout);
|
|
|
|
}
|
|
|
|
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
/*
|
|
|
|
* Find the next thread to run when there is no thread in the cache and update
|
|
|
|
* the cache.
|
|
|
|
*/
|
|
|
|
static struct k_thread *__get_next_ready_thread(void)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
|
|
|
int prio = _get_highest_ready_prio();
|
|
|
|
int q_index = _get_ready_q_q_index(prio);
|
2016-11-08 10:36:50 -05:00
|
|
|
sys_dlist_t *list = &_ready_q.q[q_index];
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
2016-09-29 11:03:51 -04:00
|
|
|
__ASSERT(!sys_dlist_is_empty(list),
|
|
|
|
"no thread to run (prio: %d, queue index: %u)!\n",
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
prio, q_index);
|
|
|
|
|
2016-09-29 11:03:51 -04:00
|
|
|
struct k_thread *thread =
|
|
|
|
(struct k_thread *)sys_dlist_peek_head_not_empty(list);
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
_ready_q.cache = thread;
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
return thread;
|
|
|
|
}
|
|
|
|
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
/* find which one is the next thread to run */
|
|
|
|
/* must be called with interrupts locked */
|
|
|
|
struct k_thread *_get_next_ready_thread(void)
|
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
struct k_thread *cache = _ready_q.cache;
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
|
|
|
return cache ? cache : __get_next_ready_thread();
|
|
|
|
}
|
|
|
|
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
/*
|
|
|
|
* Check if there is a thread of higher prio than the current one. Should only
|
|
|
|
* be called if we already know that the current thread is preemptible.
|
|
|
|
*/
|
|
|
|
int __must_switch_threads(void)
|
|
|
|
{
|
|
|
|
K_DEBUG("current prio: %d, highest prio: %d\n",
|
2016-11-18 16:08:24 -05:00
|
|
|
_current->base.prio, _get_highest_ready_prio());
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
extern void _dump_ready_q(void);
|
|
|
|
_dump_ready_q();
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
return _is_prio_higher(_get_highest_ready_prio(), _current->base.prio);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
}
|
|
|
|
|
2016-09-16 10:08:52 -04:00
|
|
|
int _is_next_thread_current(void)
|
|
|
|
{
|
|
|
|
return _get_next_ready_thread() == _current;
|
|
|
|
}
|
|
|
|
|
2016-10-07 13:41:34 -05:00
|
|
|
int k_thread_priority_get(k_tid_t thread)
|
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
return thread->base.prio;
|
2016-10-07 13:41:34 -05:00
|
|
|
}
|
|
|
|
|
2016-10-13 08:10:07 -04:00
|
|
|
void k_thread_priority_set(k_tid_t tid, int prio)
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
{
|
2016-11-08 15:44:05 -05:00
|
|
|
/*
|
|
|
|
* Use NULL, since we cannot know what the entry point is (we do not
|
|
|
|
* keep track of it) and idle cannot change its priority.
|
|
|
|
*/
|
|
|
|
_ASSERT_VALID_PRIO(prio, NULL);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
__ASSERT(!_is_in_isr(), "");
|
|
|
|
|
2016-10-13 08:10:07 -04:00
|
|
|
struct k_thread *thread = (struct k_thread *)tid;
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
int key = irq_lock();
|
|
|
|
|
|
|
|
_thread_priority_set(thread, prio);
|
|
|
|
_reschedule_threads(key);
|
|
|
|
}
|
|
|
|
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
/*
|
|
|
|
* Interrupts must be locked when calling this function.
|
|
|
|
*
|
|
|
|
* This function, along with _add_thread_to_ready_q() and
|
|
|
|
* _remove_thread_from_ready_q(), are the _only_ places where a thread is
|
|
|
|
* taken off or put on the ready queue.
|
|
|
|
*/
|
|
|
|
void _move_thread_to_end_of_prio_q(struct k_thread *thread)
|
|
|
|
{
|
2016-11-08 10:36:50 -05:00
|
|
|
int q_index = _get_ready_q_q_index(thread->base.prio);
|
|
|
|
sys_dlist_t *q = &_ready_q.q[q_index];
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
if (sys_dlist_is_tail(q, &thread->base.k_q_node)) {
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
sys_dlist_remove(&thread->base.k_q_node);
|
|
|
|
sys_dlist_append(q, &thread->base.k_q_node);
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
2016-11-08 10:36:50 -05:00
|
|
|
struct k_thread **cache = &_ready_q.cache;
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
|
|
|
|
*cache = *cache == thread ? NULL : *cache;
|
|
|
|
}
|
|
|
|
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
void k_yield(void)
|
|
|
|
{
|
|
|
|
__ASSERT(!_is_in_isr(), "");
|
|
|
|
|
|
|
|
int key = irq_lock();
|
|
|
|
|
unified: cache the next thread to run
When adding a thread to the ready queue, it is often known at that time
if the thread added will be the next one to run or not. So, instead of
simply updating the ready queues and the bitmask, also cache what that
thread is, so that when the scheduler is invoked, it can simply fetch it
from there. This is only done if there is a thread in the cache, since
the way the cache is updated is by comparing the priorities of the
thread being added and the cached thread.
When a thread is removed from the ready queue, if it is currently the
cached thread, it is also removed from the cache. The cache is not
updated at this time, since this would be a preemptive fetching that
could be overriden before the newly cached thread would even be
scheduled in.
Finally, when a thread is scheduled in, it now becomes the cached thread
since the fact that it is running means that by definition it was the
next one to run.
Doing this can speed up considerably some context switch times,
especially when a thread is preempted by an interrupt and the same
thread is scheduled when the interrupt exits.
Change-Id: I6dc8391cfca566699bb9b217eafe6bc6a063c8bb
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-30 13:44:58 -04:00
|
|
|
_move_thread_to_end_of_prio_q(_current);
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
if (_current == _get_next_ready_thread()) {
|
|
|
|
irq_unlock(key);
|
|
|
|
} else {
|
|
|
|
_Swap(key);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void k_sleep(int32_t duration)
|
|
|
|
{
|
|
|
|
__ASSERT(!_is_in_isr(), "");
|
2016-10-05 16:03:31 -04:00
|
|
|
__ASSERT(duration != K_FOREVER, "");
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
K_DEBUG("thread %p for %d ns\n", _current, duration);
|
|
|
|
|
|
|
|
/* wait of 0 ns is treated as a 'yield' */
|
|
|
|
if (duration == 0) {
|
|
|
|
k_yield();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
int key = irq_lock();
|
|
|
|
|
|
|
|
_mark_thread_as_timing(_current);
|
|
|
|
_remove_thread_from_ready_q(_current);
|
2016-10-17 14:34:53 -05:00
|
|
|
_add_thread_timeout(_current, NULL,
|
|
|
|
_TICK_ALIGN + _ms_to_ticks(duration));
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
|
|
|
|
_Swap(key);
|
|
|
|
}
|
|
|
|
|
|
|
|
void k_wakeup(k_tid_t thread)
|
|
|
|
{
|
|
|
|
int key = irq_lock();
|
|
|
|
|
|
|
|
/* verify first if thread is not waiting on an object */
|
2016-10-06 16:25:39 -04:00
|
|
|
if (_is_thread_pending(thread)) {
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
irq_unlock(key);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2016-10-05 12:55:17 -04:00
|
|
|
if (_abort_thread_timeout(thread) < 0) {
|
unified: initial unified kernel implementation
Summary of what this includes:
initialization:
Copy from nano_init.c, with the following changes:
- the main thread is the continuation of the init thread, but an idle
thread is created as well
- _main() initializes threads in groups and starts the EXE group
- the ready queues are initialized
- the main thread is marked as non-essential once the system init is
done
- a weak main() symbol is provided if the application does not provide a
main() function
scheduler:
Not an exhaustive list, but basically provide primitives for:
- adding/removing a thread to/from a wait queue
- adding/removing a thread to/from the ready queue
- marking thread as ready
- locking/unlocking the scheduler
- instead of locking interrupts
- getting/setting thread priority
- checking what state (coop/preempt) a thread is currenlty running in
- rescheduling threads
- finding what thread is the next to run
- yielding/sleeping/aborting sleep
- finding the current thread
threads:
- Add operationns on threads, such as creating and starting them.
standardized handling of kernel object return codes:
- Kernel objects now cause _Swap() to return the following values:
0 => operation successful
-EAGAIN => operation timed out
-Exxxxx => operation failed for another reason
- The thread's swap_data field can be used to return any additional
information required to complete the operation, such as the actual
result of a successful operation.
timeouts:
- same as nano timeouts, renamed to simply 'timeouts'
- the kernel is still tick-based, but objects take timeout values in
ms for forward compatibility with a tickless kernel.
semaphores:
- Port of the nanokernel semaphores, which have the same basic behaviour
as the microkernel ones. Semaphore groups are not yet implemented.
- These semaphores are enhanced in that they accept an initial count and a
count limit. This allows configuring them as binary semaphores, and also
provisioning them without having to "give" the semaphore multiple times
before using them.
mutexes:
- Straight port of the microkernel mutexes. An init function is added to
allow defining them at runtime.
pipes:
- straight port
timers:
- amalgamation of nano and micro timers, with all functionalities
intact.
events:
- re-implementation, using semaphores and workqueues.
mailboxes:
- straight port
message queues:
- straight port of microkernel FIFOs
memory maps:
- straight port
workqueues:
- Basically, have all APIs follow the k_ naming rule, and use the _timeout
subsystem from the unified kernel directory, and not the _nano_timeout
one.
stacks:
- Port of the nanokernel stacks. They can now have multiple threads
pending on them and threads can wait with a timeout.
LIFOs:
- Straight port of the nanokernel LIFOs.
FIFOs:
- Straight port of the nanokernel FIFOs.
Work by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Peter Mitsis <peter.mitsis@windriver.com>
Allan Stephens <allan.stephens@windriver.com>
Benjamin Walsh <benjamin.walsh@windriver.com>
Change-Id: Id3cadb3694484ab2ca467889cfb029be3cd3a7d6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-09-02 18:55:39 -04:00
|
|
|
irq_unlock(key);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
_ready_thread(thread);
|
|
|
|
|
|
|
|
if (_is_in_isr()) {
|
|
|
|
irq_unlock(key);
|
|
|
|
} else {
|
|
|
|
_reschedule_threads(key);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
k_tid_t k_current_get(void)
|
|
|
|
{
|
|
|
|
return _current;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* debug aid */
|
|
|
|
void _dump_ready_q(void)
|
|
|
|
{
|
|
|
|
K_DEBUG("bitmap: %x\n", _ready_q.prio_bmap[0]);
|
|
|
|
for (int prio = 0; prio < K_NUM_PRIORITIES; prio++) {
|
|
|
|
K_DEBUG("prio: %d, head: %p\n",
|
|
|
|
prio - CONFIG_NUM_COOP_PRIORITIES,
|
|
|
|
sys_dlist_peek_head(&_ready_q.q[prio]));
|
|
|
|
}
|
|
|
|
}
|
2016-09-12 11:35:26 -04:00
|
|
|
|
|
|
|
#ifdef CONFIG_TIMESLICING
|
|
|
|
extern int32_t _time_slice_duration; /* Measured in ms */
|
|
|
|
extern int32_t _time_slice_elapsed; /* Measured in ms */
|
|
|
|
extern int _time_slice_prio_ceiling;
|
|
|
|
|
|
|
|
void k_sched_time_slice_set(int32_t duration_in_ms, int prio)
|
|
|
|
{
|
|
|
|
__ASSERT(duration_in_ms >= 0, "");
|
|
|
|
__ASSERT((prio >= 0) && (prio < CONFIG_NUM_PREEMPT_PRIORITIES), "");
|
|
|
|
|
|
|
|
_time_slice_duration = duration_in_ms;
|
|
|
|
_time_slice_elapsed = 0;
|
|
|
|
_time_slice_prio_ceiling = prio;
|
|
|
|
}
|
|
|
|
#endif /* CONFIG_TIMESLICING */
|
2016-11-10 15:54:27 -05:00
|
|
|
|
|
|
|
int k_is_preempt_thread(void)
|
|
|
|
{
|
|
|
|
return !_is_in_isr() && _is_preempt(_current);
|
|
|
|
}
|