Zephyr's timeslice implementation has always been somewhat primitive.
You get a global timeslice that applies broadly to the whole bottom of
the priority space, with no ability (beyond that one priority
threshold) to tune it to work on certain threads, etc...
This adds an (optionally configurable) API that allows timeslicing to
be controlled on a per-thread basis: any thread at any priority can be
set to timeslice, for a configurable per-thread slice time, and at the
end of its slice a callback can be provided that can take action.
This allows the application to implement things like responsiveness
heuristics, "fair" scheduling algorithms, etc... without requiring
that facility in the core kernel.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Zeroing the BSS and copying data to RAM with regular memset/memcpy may
cause problems when those functions are assuming a fully initialized
system for their optimizations to work e.g. some instructions require
an active MMU, but turning the MMU on needs the .bss section to be
cleared first, etc.
Commit c5b898743a ("aarch64: Fix alignment fault on z_bss_zero()")
provides a detailed explanation of such a case.
Replacing z_bss_zero() with an architecture specific one is problematic
as the former may see new sections added to it that would be missed by
the later. The same reasoning goes for z_data_copy().
Let's make maintenance much easier by providing weak versions of
memset/memcpy that can be overridden by architecture-specific safe
versions when needed.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Extracting stack usage calculation from k_thread_stack_space_get to
z_stack_space_get so it can be used also for interrupt stack.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
There is no need to use conditional compilation for the function
prototypes in the kernel architecture header file. So remove it.
Added bouns is that these functions can appear in documentation
without explicitly enabled in pre-defines during doc build.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Extends the CPU usage runtime stats to track current, total, peak
and average usage (as bounded by the scheduling of the idle thread).
This permits a developer to obtain more system information if desired
to tune the system.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
This commit does two things to the z_sched_thread_usage(). First,
it updates the API so that it accepts a pointer to the runtime
stats instead of simply returning the usage cycles. This gives it
the flexibility to retrieve additional statistics in the future.
Second, the runtime stats are only updated if the specified thread
is the current thread running on the current core.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
The resource pool of the short-lived dummy thread "stub" may be
inherited by other threads created during system initialization. This
commit initializes this resource pool to NULL or the system pool to
ensure that a well-defined resource pool propagates to other threads
that inherit it from the dummy thread.
Fixes#41482.
Signed-off-by: Berend Ozceri <berend@recogni.com>
Storing the state where this is the first GDB break can be done
in the main GDB stub code. There is no need to store the state
in architecture layer.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Remove LOG_MINIMAL kconfig option which was confusing
since LOG_MODE_MINIMAL existed. LOG_MINIMAL was used to
force minimal mode but because of invalid dependencies
it was leading to issues.
Refactored code to use LOG_MODE_MINIMAL everywhere and
renamed LOG_MINIMAL to LOG_DEFAULT_MINIMAL which has impact
on defualt logging mode (which still can be later changed
in conf file or in menuconfig).
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Clean up RUNTIME_STATS to separate the API from the individual data
backends. Use the SCHED_THREAD_USAGE tracking instead of the original
for execution_cycles. Move the kconfig for that into the runtime
stats menu, since it's part of the family now.
Also remove a lot of needless #if's around the declarations. Unused
structs and uncalled functions don't need to be explicitly hidden. An
attempt to access a non-existent field (e.g. "execution_cycles" if
that isn't configured) provides all the build time validation we need.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This is an alternate backend that does what THREAD_RUNTIME_STATS is
doing currently, but with a few advantages:
* Correctly synchronized: you can't race against a running thread
(potentially on another CPU!) while querying its usage.
* Realtime results: you get the right answer always, up to timer
precision, even if a thread has been running for a while
uninterrupted and hasn't updated its total.
* Portable, no need for per-architecture code at all for the simple
case. (It leverages the USE_SWITCH layer to do this, so won't work
on older architectures)
* Faster/smaller: minimizes use of 64 bit math; lower overhead in
thread struct (keeps the scratch "started" time in the CPU struct
instead). One 64 bit counter per thread and a 32 bit scratch
register in the CPU struct.
* Standalone. It's a core (but optional) scheduler feature, no
dependence on para-kernel configuration like the tracing
infrastructure.
* More precise: allows architectures to optionally call a trivial
zero-argument/no-result cdecl function out of interrupt entry to
avoid accounting for ISR runtime in thread totals. No configuration
needed here, if it's called then you get proper ISR accounting, and
if not you don't.
For right now, pending unification, it's added side-by-side with the
older API and left as a z_*() internal symbol.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Instead of returning PM_STATE_ACTIVE for when the cpu didn't enter a
low power state and a different state when it entered, but has
already left the state and is active again, it changes
pm_system_suspend to return true when the cpu has entered a low power
state and false otherwise.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
There was a brief (but seen in practice on real apps on real
hardware!) race with the switch-based z_swap() implementation. The
thread return value was being initialized to -EAGAIN after the
enclosing lock had been released. But that lock is supposed to be
atomic with the thread suspend.
This opened a window for another racing thread to come by and "wake
up" our pending thread (which is fine on its own), set its return
value (e.g. to 0 for success) and then have that value clobbered by
the thread continuing to suspend itself outside the lock.
Melodramatic aside: I continue to hate this
arch_thread_return_value_set() API; it needs to die. At best it's a
mild optimization on a handful of architectures (e.g. x86 implements
it by writing to the EAX register save slot in the context block).
Asynchronous APIs are almost always worse than synchronous ones, and
in this case it's an async operation that races against literal
context switch code that can't use traditional locking strategies.
Fixes#39575
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Some SMP applications have threading designs where every thread
created is always assigned to a specific CPU, and never want to
schedule them symmetrically across CPUs under any circumstance.
In this situation, it's possible to optimize the run queue design a
bit to put a separate queue in each CPU struct instead of having a
single global one. This is probably good for a few cycles per
scheduling event (maybe a bit more on architectures where cache
locality can be exploited) in circumstances where there is more than
one runnable thread. It's a mild optimization, but a basically simple
one.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Some architectures already returns -ENOTSUP when these functions
are called. So add this return value to the API doc.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Add a SOC API to allow for application control over deep idle power
states. Note that the hardware idle entry happens out of the WAITI
instruction, so the application has to be responsibile for ensuring
the CPU to be halted actually reaches idle deterministically. Lots of
warnings in the docs to this effect.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
During boot process, the boot sections need to be pinned in
memory to prevent them from being paged out (to avoid
pages being paged out and immediately paged in again).
Once the boot process is completed (just before calling main()),
the boot sections can be unpinned so the memory can be
used for demand paging for paging in data pages.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
z_smp_init() is only available if CONFIG_SMP is defined,
smp_timer_init() also depends on two Kconfig parameters. Also make it
conditional in cavs_timer.c. Also clarify some SMP-related comments
there.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
The z_interrupt_stacks was declared extern in the kernel internal
header file using the same macro which defines the same stack
array but with an added "extern" in front. This macro adds
alignment and section attribute which are actually not the same
as the actual stack array defined in kernel/init.c. The section
name used in the section attribute contains the file name where
the stack array is defined or extern declared. So the same
symbol, in this case z_interrupt_stacks, has different
attributes in two places, and GCC 11 starts to complain about
this. So use the newly introduced macro to extern declare
the stack array without adding/replacing any symbol attributes.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
These functions are those that need be implemented by backing
store outside kernel. Promote them from z_* so these can be
included in documentation.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
These functions and data structures are those that need
to be implemented by eviction algorithm and application
outside kernel. Promote them from z_* so these can be
included in documentation.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The scheduler has historically had an API where an application can
inform the kernel that it will never create a thread that can be
preempted, and the kernel and architecture layer would use that as an
optimization hint to eliminate some code paths.
Those optimizations have dwindled to almost nothing at this point, and
they're now objectively a smaller impact than the special casing that
was required to handle the idle thread (which, obviously, must always
be preemptible).
Fix this by eliminating the idea of "cooperative only" and ensuring
that there will always be at least one preemptible priority with value
>=0. CONFIG_NUM_PREEMPT_PRIORITIES now specifies the number of
user-accessible priorities other than the idle thread.
The only remaining workaround is that some older architectures (and
also SPARC) use the CONFIG_PREEMPT_ENABLED=n state as a hint to skip
thread switching on interrupt exit. So detect exactly those platforms
and implement a minimal workaround in the idle loop (basically "just
call swap()") instead, with a big explanation.
Note that this also fixes a bug in one of the philosophers samples,
where it would ask for 6 cooperative priorities but then use values -7
through -2. It was assuming the kernel would magically create a
cooperative priority for its idle thread, which wasn't correct even
before.
Fixes#34584
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Our z_swap() API takes a key returned from arch_irq_lock() and
releases it atomically with the context switch. Make sure that the
action of the unlocking is to unmask interrupts globally. If
interrupts would still be masked then that means there is an OUTER
interrupt lock still held, and the code that locked it surely doesn't
expect the thread to be suspended and interrupts unmasked while it's
held!
Unfortunately, this kind of mistake is very easy to make. We should
catch that with a simple assertion. This is essentially a crude
Zephyr equivalent of the extremely common "BUG: scheduling while
atomic" error in Linux drivers (just google it).
The one exception made is the circumstance where a thread has already
aborted itself. At that stage, whatever upthread lock state might
have existed will have already been messed up, so there's no value in
our asserting here. We can't catch all bugs, and this can actually
happen in error handling and/or test frameworks.
Fixes#33319
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
