github
/
yuzu
mirror of https://github.com/yuzu-emu/yuzu
1
Fork 0
Code Releases Activity
yuzu/src/core/hle/kernel/kernel.cpp

286 lines
9.3 KiB
C++
Raw Normal View History

License change
2014-12-17 06:38:14 +01:00
// Copyright 2014 Citra Emulator Project
// Licensed under GPLv2 or any later version
Remove trailing spaces in every file but the ones imported from SkyEye, AOSP or generated
2014-11-19 09:49:13 +01:00
// Refer to the license.txt file included.
added initial kernel/thread modules
2014-05-10 04:11:18 +02:00
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
#include <atomic>
#include <memory>
#include <mutex>
#include <utility>
#include "common/assert.h"
#include "common/logging/log.h"
#include "core/core.h"
#include "core/core_timing.h"
Kernel: Initial implementation of thread preemption.
2019-09-10 17:04:40 +02:00
#include "core/core_timing_util.h"
kernel/address_arbiter: Convert the address arbiter into a class Places all of the functions for address arbiter operation into a class. This will be necessary for future deglobalizing efforts related to both the memory and system itself.
2019-03-05 17:54:06 +01:00
#include "core/hle/kernel/address_arbiter.h"
service: Migrate global named port map to the KernelCore class Now that we have a class representing the kernel in some capacity, we now have a place to put the named port map, so we move it over and get rid of another piece of global state within the core.
2018-09-02 17:58:58 +02:00
#include "core/hle/kernel/client_port.h"
Kernel: Correct Results in Condition Variables and Mutexes
2019-10-08 00:57:13 +02:00
#include "core/hle/kernel/errors.h"
Kernel: Move HandleTable to a separate file
2017-05-30 01:45:42 +02:00
#include "core/hle/kernel/handle_table.h"
Use negative priorities to avoid special-casing the self-include
2016-09-21 08:52:38 +02:00
#include "core/hle/kernel/kernel.h"
Kernel: Introduce skeleton Process class to hold process data
2015-05-04 05:01:16 +02:00
#include "core/hle/kernel/process.h"
Kernel: Add more infrastructure to support different memory layouts This adds some structures necessary to support multiple memory regions in the future. It also adds support for different system memory types and the new linear heap mapping at 0x30000000.
2015-08-06 02:26:52 +02:00
#include "core/hle/kernel/resource_limit.h"
Add interfacing to the Global Scheduler
2019-03-29 22:02:57 +01:00
#include "core/hle/kernel/scheduler.h"
added initial kernel/thread modules
2014-05-10 04:11:18 +02:00
#include "core/hle/kernel/thread.h"
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
#include "core/hle/lock.h"
#include "core/hle/result.h"
kernel: Handle page table switching within MakeCurrentProcess() Centralizes the page table switching to one spot, rather than making calling code deal with it everywhere.
2019-04-07 07:10:44 +02:00
#include "core/memory.h"
added initial kernel/thread modules
2014-05-10 04:11:18 +02:00
- created a Kernel namespace - cleaned up Kernel code a bit (moved stuff into namespace, fixed whitespace issues) - added handle types for all different CTROS handles
2014-05-21 00:13:25 +02:00
namespace Kernel {
added initial kernel/thread modules
2014-05-10 04:11:18 +02:00
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
/**
* Callback that will wake up the thread it was scheduled for
* @param thread_handle The handle of the thread that's been awoken
* @param cycles_late The number of CPU cycles that have passed since the desired wakeup time
*/
core/core_timing: Make callback parameters consistent In some cases, our callbacks were using s64 as a parameter, and in other cases, they were using an int, which is inconsistent. To make all callbacks consistent, we can just use an s64 as the type for late cycles, given it gets rid of the need to cast internally. While we're at it, also resolve some signed/unsigned conversions that were occurring related to the callback registration.
2019-03-24 23:11:32 +01:00
static void ThreadWakeupCallback(u64 thread_handle, [[maybe_unused]] s64 cycles_late) {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
const auto proper_handle = static_cast<Handle>(thread_handle);
core: Make System references const where applicable
2018-10-28 22:44:46 +01:00
const auto& system = Core::System::GetInstance();
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// Lock the global kernel mutex when we enter the kernel HLE.
general: Use deducation guides for std::lock_guard and std::unique_lock Since C++17, the introduction of deduction guides for locking facilities means that we no longer need to hardcode the mutex type into the locks themselves, making it easier to switch mutex types, should it ever be necessary in the future.
2019-04-01 18:29:59 +02:00
std::lock_guard lock{HLE::g_hle_lock};
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
std::shared_ptr<Thread> thread =
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
system.Kernel().RetrieveThreadFromWakeupCallbackHandleTable(proper_handle);
if (thread == nullptr) {
LOG_CRITICAL(Kernel, "Callback fired for invalid thread {:08X}", proper_handle);
return;
}
bool resume = true;
kernel/thread: Unify wait synchronization types This is a holdover from Citra, where the 3DS has both WaitSynchronization1 and WaitSynchronizationN. The switch only has one form of wait synchronizing (literally WaitSynchonization). This allows us to throw out code that doesn't apply at all to the Switch kernel. Because of this unnecessary dichotomy within the wait synchronization utilities, we were also neglecting to properly handle waiting on multiple objects. While we're at it, we can also scrub out any lingering references to WaitSynchronization1/WaitSynchronizationN in comments, and change them to WaitSynchronization (or remove them if the mention no longer applies).
2019-04-17 13:21:19 +02:00
if (thread->GetStatus() == ThreadStatus::WaitSynch ||
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
thread->GetStatus() == ThreadStatus::WaitHLEEvent) {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// Remove the thread from each of its waiting objects' waitlists
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
for (const auto& object : thread->GetWaitObjects()) {
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
object->RemoveWaitingThread(thread);
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
thread->ClearWaitObjects();
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// Invoke the wakeup callback before clearing the wait objects
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
if (thread->HasWakeupCallback()) {
resume = thread->InvokeWakeupCallback(ThreadWakeupReason::Timeout, thread, nullptr, 0);
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
Kernel: Correct Results in Condition Variables and Mutexes
2019-10-08 00:57:13 +02:00
} else if (thread->GetStatus() == ThreadStatus::WaitMutex ||
thread->GetStatus() == ThreadStatus::WaitCondVar) {
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
thread->SetMutexWaitAddress(0);
thread->SetWaitHandle(0);
Kernel: Correct behavior of Condition Variables to be more similar to real hardware. This commit ensures cond var threads act exactly as they do in the real console. The original implementation uses an RBTree and the behavior of cond var threads is that at the same priority level they act like a FIFO.
2019-11-15 01:13:18 +01:00
if (thread->GetStatus() == ThreadStatus::WaitCondVar) {
thread->GetOwnerProcess()->RemoveConditionVariableThread(thread);
Kernel: Optimize condition variable threads management.
2019-11-21 16:03:37 +01:00
thread->SetCondVarWaitAddress(0);
Kernel: Correct behavior of Condition Variables to be more similar to real hardware. This commit ensures cond var threads act exactly as they do in the real console. The original implementation uses an RBTree and the behavior of cond var threads is that at the same priority level they act like a FIFO.
2019-11-15 01:13:18 +01:00
}
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
auto* const lock_owner = thread->GetLockOwner();
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// Threads waking up by timeout from WaitProcessWideKey do not perform priority inheritance
// and don't have a lock owner unless SignalProcessWideKey was called first and the thread
// wasn't awakened due to the mutex already being acquired.
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
if (lock_owner != nullptr) {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
lock_owner->RemoveMutexWaiter(thread);
}
}
kernel/thread: Make all instance variables private Many of the member variables of the thread class aren't even used outside of the class itself, so there's no need to make those variables public. This change follows in the steps of the previous changes that made other kernel types' members private. The main motivation behind this is that the Thread class will likely change in the future as emulation becomes more accurate, and letting random bits of the emulator access data members of the Thread class directly makes it a pain to shuffle around and/or modify internals. Having all data members public like this also makes it difficult to reason about certain bits of behavior without first verifying what parts of the core actually use them. Everything being public also generally follows the tendency for changes to be introduced in completely different translation units that would otherwise be better introduced as an addition to the Thread class' public interface.
2018-10-04 00:47:57 +02:00
if (thread->GetArbiterWaitAddress() != 0) {
ASSERT(thread->GetStatus() == ThreadStatus::WaitArb);
thread->SetArbiterWaitAddress(0);
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
if (resume) {
Kernel: Correct Results in Condition Variables and Mutexes
2019-10-08 00:57:13 +02:00
if (thread->GetStatus() == ThreadStatus::WaitCondVar ||
thread->GetStatus() == ThreadStatus::WaitArb) {
thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
}
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
thread->ResumeFromWait();
}
}
struct KernelCore::Impl {
Kernel: Style and Corrections
2019-06-19 15:11:18 +02:00
explicit Impl(Core::System& system) : system{system}, global_scheduler{system} {}
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
void Initialize(KernelCore& kernel) {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
Shutdown();
Kernel: Properly initialize and shutdown all modules.
2015-04-28 04:12:35 +02:00
kernel/resource_limit: Clean up interface Cleans out the citra/3DS-specific implementation details that don't apply to the Switch. Sets the stage for implementing ResourceLimit instances properly. While we're at it, remove the erroneous checks within CreateThread() and SetThreadPriority(). While these are indeed checked in some capacity, they are not checked via a ResourceLimit instance. In the process of moving out Citra-specifics, this also replaces the system ResourceLimit instance's values with ones from the Switch.
2018-11-19 18:54:06 +01:00
InitializeSystemResourceLimit(kernel);
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
InitializeThreads();
Kernel: Initial implementation of thread preemption.
2019-09-10 17:04:40 +02:00
InitializePreemption();
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
void Shutdown() {
next_object_id = 0;
kernel: Differentiate kernel and user processes when picking ID This allows kernel internal type processes to be assigned IDs in the KIP range while userland processes are assigned in the user range.
2019-06-10 06:28:33 +02:00
next_kernel_process_id = Process::InitialKIPIDMin;
next_user_process_id = Process::ProcessIDMin;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
next_thread_id = 1;
process_list.clear();
kernel/thread: Use a regular pointer for the owner/current process There's no real need to use a shared pointer in these cases, and only makes object management more fragile in terms of how easy it would be to introduce cycles. Instead, just do the simple thing of using a regular pointer. Much of this is just a hold-over from citra anyways. It also doesn't make sense from a behavioral point of view for a process' thread to prolong the lifetime of the process itself (the process is supposed to own the thread, not the other way around).
2018-10-10 06:42:10 +02:00
current_process = nullptr;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
kernel/resource_limit: Clean up interface Cleans out the citra/3DS-specific implementation details that don't apply to the Switch. Sets the stage for implementing ResourceLimit instances properly. While we're at it, remove the erroneous checks within CreateThread() and SetThreadPriority(). While these are indeed checked in some capacity, they are not checked via a ResourceLimit instance. In the process of moving out Citra-specifics, this also replaces the system ResourceLimit instance's values with ones from the Switch.
2018-11-19 18:54:06 +01:00
system_resource_limit = nullptr;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
thread_wakeup_callback_handle_table.Clear();
thread_wakeup_event_type = nullptr;
Kernel: Initial implementation of thread preemption.
2019-09-10 17:04:40 +02:00
preemption_event = nullptr;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
Kernel Scheduler: Make sure the global scheduler shutdowns correctly.
2019-10-12 14:21:51 +02:00
global_scheduler.Shutdown();
service: Migrate global named port map to the KernelCore class Now that we have a class representing the kernel in some capacity, we now have a place to put the named port map, so we move it over and get rid of another piece of global state within the core.
2018-09-02 17:58:58 +02:00
named_ports.clear();
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
kernel/resource_limit: Clean up interface Cleans out the citra/3DS-specific implementation details that don't apply to the Switch. Sets the stage for implementing ResourceLimit instances properly. While we're at it, remove the erroneous checks within CreateThread() and SetThreadPriority(). While these are indeed checked in some capacity, they are not checked via a ResourceLimit instance. In the process of moving out Citra-specifics, this also replaces the system ResourceLimit instance's values with ones from the Switch.
2018-11-19 18:54:06 +01:00
// Creates the default system resource limit
void InitializeSystemResourceLimit(KernelCore& kernel) {
kernel/resource_limit: Remove the name member from resource limits This doesn't really provide any benefit to the resource limit interface. There's no way for callers to any of the service functions for resource limits to provide a custom name, so all created instances of resource limits other than the system resource limit would have a name of "Unknown". The system resource limit itself is already trivially identifiable from its limit values, so there's no real need to take up space in the object to identify one object meaningfully out of N total objects.
2019-04-01 22:46:00 +02:00
system_resource_limit = ResourceLimit::Create(kernel);
kernel/resource_limit: Clean up interface Cleans out the citra/3DS-specific implementation details that don't apply to the Switch. Sets the stage for implementing ResourceLimit instances properly. While we're at it, remove the erroneous checks within CreateThread() and SetThreadPriority(). While these are indeed checked in some capacity, they are not checked via a ResourceLimit instance. In the process of moving out Citra-specifics, this also replaces the system ResourceLimit instance's values with ones from the Switch.
2018-11-19 18:54:06 +01:00
// If setting the default system values fails, then something seriously wrong has occurred.
ASSERT(system_resource_limit->SetLimitValue(ResourceType::PhysicalMemory, 0x200000000)
.IsSuccess());
ASSERT(system_resource_limit->SetLimitValue(ResourceType::Threads, 800).IsSuccess());
ASSERT(system_resource_limit->SetLimitValue(ResourceType::Events, 700).IsSuccess());
ASSERT(system_resource_limit->SetLimitValue(ResourceType::TransferMemory, 200).IsSuccess());
ASSERT(system_resource_limit->SetLimitValue(ResourceType::Sessions, 900).IsSuccess());
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
void InitializeThreads() {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
thread_wakeup_event_type =
core_timing: Use better reference tracking for EventType. (#3159) * core_timing: Use better reference tracking for EventType. - Moves ownership of the event to the caller, ensuring we don't fire events for destroyed objects. - Removes need for unique names - we won't be using this for save states anyways.
2019-11-27 03:48:56 +01:00
Core::Timing::CreateEvent("ThreadWakeupCallback", ThreadWakeupCallback);
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
Kernel: Initial implementation of thread preemption.
2019-09-10 17:04:40 +02:00
void InitializePreemption() {
core_timing: Use better reference tracking for EventType. (#3159) * core_timing: Use better reference tracking for EventType. - Moves ownership of the event to the caller, ensuring we don't fire events for destroyed objects. - Removes need for unique names - we won't be using this for save states anyways.
2019-11-27 03:48:56 +01:00
preemption_event =
Core::Timing::CreateEvent("PreemptionCallback", [this](u64 userdata, s64 cycles_late) {
Kernel: Initial implementation of thread preemption.
2019-09-10 17:04:40 +02:00
global_scheduler.PreemptThreads();
s64 time_interval = Core::Timing::msToCycles(std::chrono::milliseconds(10));
system.CoreTiming().ScheduleEvent(time_interval, preemption_event);
});
s64 time_interval = Core::Timing::msToCycles(std::chrono::milliseconds(10));
system.CoreTiming().ScheduleEvent(time_interval, preemption_event);
}
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
std::atomic<u32> next_object_id{0};
kernel: Differentiate kernel and user processes when picking ID This allows kernel internal type processes to be assigned IDs in the KIP range while userland processes are assigned in the user range.
2019-06-10 06:28:33 +02:00
std::atomic<u64> next_kernel_process_id{Process::InitialKIPIDMin};
std::atomic<u64> next_user_process_id{Process::ProcessIDMin};
kernel/thread: Make thread_id a 64-bit value The kernel uses a 64-bit value for the thread ID, so we shouldn't be using a 32-bit value.
2018-12-19 04:37:01 +01:00
std::atomic<u64> next_thread_id{1};
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// Lists all processes that exist in the current session.
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
std::vector<std::shared_ptr<Process>> process_list;
kernel/thread: Use a regular pointer for the owner/current process There's no real need to use a shared pointer in these cases, and only makes object management more fragile in terms of how easy it would be to introduce cycles. Instead, just do the simple thing of using a regular pointer. Much of this is just a hold-over from citra anyways. It also doesn't make sense from a behavioral point of view for a process' thread to prolong the lifetime of the process itself (the process is supposed to own the thread, not the other way around).
2018-10-10 06:42:10 +02:00
Process* current_process = nullptr;
Add interfacing to the Global Scheduler
2019-03-29 22:02:57 +01:00
Kernel::GlobalScheduler global_scheduler;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
std::shared_ptr<ResourceLimit> system_resource_limit;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
core_timing: Use better reference tracking for EventType. (#3159) * core_timing: Use better reference tracking for EventType. - Moves ownership of the event to the caller, ensuring we don't fire events for destroyed objects. - Removes need for unique names - we won't be using this for save states anyways.
2019-11-27 03:48:56 +01:00
std::shared_ptr<Core::Timing::EventType> thread_wakeup_event_type;
std::shared_ptr<Core::Timing::EventType> preemption_event;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
// TODO(yuriks): This can be removed if Thread objects are explicitly pooled in the future,
// allowing us to simply use a pool index or similar.
Kernel::HandleTable thread_wakeup_callback_handle_table;
service: Migrate global named port map to the KernelCore class Now that we have a class representing the kernel in some capacity, we now have a place to put the named port map, so we move it over and get rid of another piece of global state within the core.
2018-09-02 17:58:58 +02:00
/// Map of named ports managed by the kernel, which can be retrieved using
/// the ConnectToPort SVC.
NamedPortTable named_ports;
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
// System context
Core::System& system;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
};
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
KernelCore::KernelCore(Core::System& system) : impl{std::make_unique<Impl>(system)} {}
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
KernelCore::~KernelCore() {
Shutdown();
}
kernel/address_arbiter: Pass in system instance to constructor Allows getting rid of reliance on the global accessor functions and instead operating on the provided system instance.
2019-03-05 18:28:10 +01:00
void KernelCore::Initialize() {
impl->Initialize(*this);
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
void KernelCore::Shutdown() {
impl->Shutdown();
}
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
std::shared_ptr<ResourceLimit> KernelCore::GetSystemResourceLimit() const {
kernel/resource_limit: Clean up interface Cleans out the citra/3DS-specific implementation details that don't apply to the Switch. Sets the stage for implementing ResourceLimit instances properly. While we're at it, remove the erroneous checks within CreateThread() and SetThreadPriority(). While these are indeed checked in some capacity, they are not checked via a ResourceLimit instance. In the process of moving out Citra-specifics, this also replaces the system ResourceLimit instance's values with ones from the Switch.
2018-11-19 18:54:06 +01:00
return impl->system_resource_limit;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
std::shared_ptr<Thread> KernelCore::RetrieveThreadFromWakeupCallbackHandleTable(
Handle handle) const {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
return impl->thread_wakeup_callback_handle_table.Get<Thread>(handle);
}
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
void KernelCore::AppendNewProcess(std::shared_ptr<Process> process) {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
impl->process_list.push_back(std::move(process));
}
kernel/thread: Use a regular pointer for the owner/current process There's no real need to use a shared pointer in these cases, and only makes object management more fragile in terms of how easy it would be to introduce cycles. Instead, just do the simple thing of using a regular pointer. Much of this is just a hold-over from citra anyways. It also doesn't make sense from a behavioral point of view for a process' thread to prolong the lifetime of the process itself (the process is supposed to own the thread, not the other way around).
2018-10-10 06:42:10 +02:00
void KernelCore::MakeCurrentProcess(Process* process) {
impl->current_process = process;
core/cpu_core_manager: Create threads separately from initialization. Our initialization process is a little wonky than one would expect when it comes to code flow. We initialize the CPU last, as opposed to hardware, where the CPU obviously needs to be first, otherwise nothing else would work, and we have code that adds checks to get around this. For example, in the page table setting code, we check to see if the system is turned on before we even notify the CPU instances of a page table switch. This results in dead code (at the moment), because the only time a page table switch will occur is when the system is *not* running, preventing the emulated CPU instances from being notified of a page table switch in a convenient manner (technically the code path could be taken, but we don't emulate the process creation svc handlers yet). This moves the threads creation into its own member function of the core manager and restores a little order (and predictability) to our initialization process. Previously, in the multi-threaded cases, we'd kick off several threads before even the main kernel process was created and ready to execute (gross!). Now the initialization process is like so: Initialization: 1. Timers 2. CPU 3. Kernel 4. Filesystem stuff (kind of gross, but can be amended trivially) 5. Applet stuff (ditto in terms of being kind of gross) 6. Main process (will be moved into the loading step in a following change) 7. Telemetry (this should be initialized last in the future). 8. Services (4 and 5 should ideally be alongside this). 9. GDB (gross. Uses namespace scope state. Needs to be refactored into a class or booted altogether). 10. Renderer 11. GPU (will also have its threads created in a separate step in a following change). Which... isn't *ideal* per-se, however getting rid of the wonky intertwining of CPU state initialization out of this mix gets rid of most of the footguns when it comes to our initialization process.
2019-04-09 19:25:54 +02:00
if (process == nullptr) {
return;
}
Memory::SetCurrentPageTable(*process);
core: Migrate current_process pointer to the kernel Given we now have the kernel as a class, it doesn't make sense to keep the current process pointer within the System class, as processes are related to the kernel. This also gets rid of a subtle case where memory wouldn't be freed on core shutdown, as the current_process pointer would never be reset, causing the pointed to contents to continue to live.
2018-09-07 02:34:51 +02:00
}
kernel/thread: Use a regular pointer for the owner/current process There's no real need to use a shared pointer in these cases, and only makes object management more fragile in terms of how easy it would be to introduce cycles. Instead, just do the simple thing of using a regular pointer. Much of this is just a hold-over from citra anyways. It also doesn't make sense from a behavioral point of view for a process' thread to prolong the lifetime of the process itself (the process is supposed to own the thread, not the other way around).
2018-10-10 06:42:10 +02:00
Process* KernelCore::CurrentProcess() {
core: Migrate current_process pointer to the kernel Given we now have the kernel as a class, it doesn't make sense to keep the current process pointer within the System class, as processes are related to the kernel. This also gets rid of a subtle case where memory wouldn't be freed on core shutdown, as the current_process pointer would never be reset, causing the pointed to contents to continue to live.
2018-09-07 02:34:51 +02:00
return impl->current_process;
}
kernel/thread: Use a regular pointer for the owner/current process There's no real need to use a shared pointer in these cases, and only makes object management more fragile in terms of how easy it would be to introduce cycles. Instead, just do the simple thing of using a regular pointer. Much of this is just a hold-over from citra anyways. It also doesn't make sense from a behavioral point of view for a process' thread to prolong the lifetime of the process itself (the process is supposed to own the thread, not the other way around).
2018-10-10 06:42:10 +02:00
const Process* KernelCore::CurrentProcess() const {
core: Migrate current_process pointer to the kernel Given we now have the kernel as a class, it doesn't make sense to keep the current process pointer within the System class, as processes are related to the kernel. This also gets rid of a subtle case where memory wouldn't be freed on core shutdown, as the current_process pointer would never be reset, causing the pointed to contents to continue to live.
2018-09-07 02:34:51 +02:00
return impl->current_process;
}
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
const std::vector<std::shared_ptr<Process>>& KernelCore::GetProcessList() const {
kernel/svc: Implement svcGetProcessList This service function simply copies out a specified number of kernel process IDs, while simultaneously reporting the total number of processes.
2019-03-20 20:03:52 +01:00
return impl->process_list;
}
Add interfacing to the Global Scheduler
2019-03-29 22:02:57 +01:00
Kernel::GlobalScheduler& KernelCore::GlobalScheduler() {
return impl->global_scheduler;
}
const Kernel::GlobalScheduler& KernelCore::GlobalScheduler() const {
return impl->global_scheduler;
}
kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. (#3154) * kernel: Replace usage of boost::intrusive_ptr with std::shared_ptr for kernel objects. - See https://github.com/citra-emu/citra/pull/4710 for details.
2019-11-25 02:15:51 +01:00
void KernelCore::AddNamedPort(std::string name, std::shared_ptr<ClientPort> port) {
service: Migrate global named port map to the KernelCore class Now that we have a class representing the kernel in some capacity, we now have a place to put the named port map, so we move it over and get rid of another piece of global state within the core.
2018-09-02 17:58:58 +02:00
impl->named_ports.emplace(std::move(name), std::move(port));
}
KernelCore::NamedPortTable::iterator KernelCore::FindNamedPort(const std::string& name) {
return impl->named_ports.find(name);
}
KernelCore::NamedPortTable::const_iterator KernelCore::FindNamedPort(
const std::string& name) const {
return impl->named_ports.find(name);
}
bool KernelCore::IsValidNamedPort(NamedPortTable::const_iterator port) const {
return port != impl->named_ports.cend();
}
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
u32 KernelCore::CreateNewObjectID() {
return impl->next_object_id++;
}
kernel/thread: Make thread_id a 64-bit value The kernel uses a 64-bit value for the thread ID, so we shouldn't be using a 32-bit value.
2018-12-19 04:37:01 +01:00
u64 KernelCore::CreateNewThreadID() {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
return impl->next_thread_id++;
- created a Kernel namespace - cleaned up Kernel code a bit (moved stuff into namespace, fixed whitespace issues) - added handle types for all different CTROS handles
2014-05-21 00:13:25 +02:00
}
kernel: Differentiate kernel and user processes when picking ID This allows kernel internal type processes to be assigned IDs in the KIP range while userland processes are assigned in the user range.
2019-06-10 06:28:33 +02:00
u64 KernelCore::CreateNewKernelProcessID() {
return impl->next_kernel_process_id++;
}
u64 KernelCore::CreateNewUserProcessID() {
return impl->next_user_process_id++;
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
}
Kernel: Add more infrastructure to support different memory layouts This adds some structures necessary to support multiple memory regions in the future. It also adds support for different system memory types and the new linear heap mapping at 0x30000000.
2015-08-06 02:26:52 +02:00
core_timing: Use better reference tracking for EventType. (#3159) * core_timing: Use better reference tracking for EventType. - Moves ownership of the event to the caller, ensuring we don't fire events for destroyed objects. - Removes need for unique names - we won't be using this for save states anyways.
2019-11-27 03:48:56 +01:00
const std::shared_ptr<Core::Timing::EventType>& KernelCore::ThreadWakeupCallbackEventType() const {
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
return impl->thread_wakeup_event_type;
}
Kernel::HandleTable& KernelCore::ThreadWakeupCallbackHandleTable() {
return impl->thread_wakeup_callback_handle_table;
}
Kernel: Add more infrastructure to support different memory layouts This adds some structures necessary to support multiple memory regions in the future. It also adds support for different system memory types and the new linear heap mapping at 0x30000000.
2015-08-06 02:26:52 +02:00
kernel: Eliminate kernel global state As means to pave the way for getting rid of global state within core, This eliminates kernel global state by removing all globals. Instead this introduces a KernelCore class which acts as a kernel instance. This instance lives in the System class, which keeps its lifetime contained to the lifetime of the System class. This also forces the kernel types to actually interact with the main kernel instance itself instead of having transient kernel state placed all over several translation units, keeping everything together. It also has a nice consequence of making dependencies much more explicit. This also makes our initialization a tad bit more correct. Previously we were creating a kernel process before the actual kernel was initialized, which doesn't really make much sense. The KernelCore class itself follows the PImpl idiom, which allows keeping all the implementation details sealed away from everything else, which forces the use of the exposed API and allows us to avoid any unnecessary inclusions within the main kernel header.
2018-08-28 18:30:33 +02:00
const Kernel::HandleTable& KernelCore::ThreadWakeupCallbackHandleTable() const {
return impl->thread_wakeup_callback_handle_table;
- added __KernelLoadExec function - fixed some logging
2014-05-14 03:57:12 +02:00
}
kernel: refactored function naming to remove "__" prefix
2014-05-23 01:06:12 +02:00
kernel: Add ObjectAddressTable class.
2018-01-01 19:25:37 +01:00
} // namespace Kernel