Commit Graph

114 Commits

Author SHA1 Message Date
bunnei
6e26d063a5 Merge pull request #757 from Subv/scheduling
Core/Scheduling: Prepare the new priority in the thread queue when svcSetPriority is called
2015-05-11 22:54:00 -04:00
Yuri Kunde Schlesner
820b97787c Merge pull request #750 from Subv/process_svc
Core/HLE: Implemented the SVCs GetProcessId and GetProcessIdOfThread
2015-05-11 17:44:26 -07:00
Subv
dbc1320923 Core/Scheduling: Prepare the new priority in the thread queue when svcSetPriority is called 2015-05-11 19:38:10 -05:00
Subv
41f74a16fd Core/HLE: Implemented the SVCs GetProcessId and GetProcessIdOfThread 2015-05-11 09:15:10 -05:00
Yuri Kunde Schlesner
d16c2bd956 Thread: Correctly set main thread initial stack position 2015-05-11 05:08:47 -03:00
Subv
115ad8e16a fixup! Set the TLS address in the scheduler 2015-05-10 18:43:59 -05:00
Subv
000876858d Core/Memory: Give every emulated thread it's own TLS area.
The TLS area for thread T with id Ti is located at TLS_AREA_VADDR + (Ti - 1) * 0x200.
This allows some games like Mario Kart 7 to continue further.
2015-05-10 18:35:37 -05:00
Yuri Kunde Schlesner
1c0b87edc2 Memory: Re-organize and rename memory area address constants 2015-05-09 01:29:52 -03:00
Yuri Kunde Schlesner
e1fbac3ca1 Common: Remove common.h 2015-05-07 15:45:22 -03:00
bunnei
c7dc799e19 Kernel: Properly initialize and shutdown all modules. 2015-05-01 18:27:03 -04:00
bunnei
9c3419ebcc Kernel: Implemented priority inheritance for mutexes. 2015-04-09 19:06:39 -04:00
bunnei
7b9f428b23 Thread: Implement priority boost for starved threads.
SVC: Return correct error code on invalid CreateThread processor ID.

SVC: Assert when creating a thread with an invalid userland priority.
2015-04-09 19:05:21 -04:00
archshift
4fb75d220a Misc cleanup of common and related functions 2015-02-19 22:26:25 -08:00
archshift
ef24e72b26 Asserts: break/crash program, fit to style guide; log.h->assert.h
Involves making asserts use printf instead of the log functions (log functions are asynchronous and, as such, the log won't be printed in time)
As such, the log type argument was removed (printf obviously can't use it, and it's made obsolete by the file and line printing)

Also removed some GEKKO cruft.
2015-02-10 18:30:31 -08:00
Kevin Hartman
5fcbfc06eb Scheduler refactor Pt. 1
* Simplifies scheduling logic, specifically regarding thread status. It should be much clearer which statuses are valid
for a thread at any given point in the system.
* Removes dead code from thread.cpp.
* Moves the implementation of resetting a ThreadContext to the corresponding core's implementation.

Other changes:
* Fixed comments in arm interfaces.
* Updated comments in thread.cpp
* Removed confusing, useless, functions like MakeReady() and ChangeStatus() from thread.cpp.
* Removed stack_size from Thread. In the CTR kernel, the thread's stack would be allocated before thread creation.
2015-02-09 21:47:12 -08:00
Yuri Kunde Schlesner
88a4a808c6 Kernel: Stop creating useless Handles during object creation
They're finally unnecessary, and will stop cluttering the application's
handle table.
2015-02-02 15:37:09 -02:00
Yuri Kunde Schlesner
52f58e64ef Kernel: Make WaitObjects share ownership of Threads waiting on them
During normal operation, a thread waiting on an WaitObject and the
object hold mutual references to each other for the duration of the
wait.

If a process is forcefully terminated (The CTR kernel has a SVC to do
this, TerminateProcess, though no equivalent exists for threads.) its
threads would also be stopped and destroyed, leaving dangling pointers
in the WaitObjects.

The solution is to simply have the Thread remove itself from WaitObjects
when it is stopped. The vector of Threads in WaitObject has also been
changed to hold SharedPtrs, just in case. (Better to have a reference
cycle than a crash.)
2015-02-02 15:37:08 -02:00
Yuri Kunde Schlesner
7725256f64 Explicitly instantiate constructors/destructors for Kernel objects
This should speed up compile times a bit, as well as enable more liberal
use of forward declarations. (Due to SharedPtr not trying to emit the
destructor anymore.)
2015-02-02 15:37:07 -02:00
Yuri Kunde Schlesner
4e84df8be3 Mutex: Replace g_mutex_held_locks with a set inside Thread 2015-02-02 15:37:06 -02:00
Yuri Kunde Schlesner
869ec46683 Kernel: Introduce unique Object ids for debugging 2015-02-02 15:37:03 -02:00
Yuri Kunde Schlesner
a9b86db3cf Kernel: Use separate Handle tables for CoreTiming userdata
This is to support the removal of GetHandle soon
2015-02-02 15:37:03 -02:00
Yuri Kunde Schlesner
664c79ff47 Thread: Modernize two functions that slipped through previous rebases 2015-02-02 15:37:01 -02:00
Lioncash
3f00dd9117 arm: Clean up ARMul_State
Remove unnecessary/unused struct variables.
2015-01-31 21:55:34 -05:00
bunnei
68ddaaa2f5 Thread: Fix WaitSynchronization1 to not set register 1 on thread wakeup. 2015-01-21 20:48:46 -05:00
bunnei
4255f25647 Thread: Use std::find in CheckWait_WaitObject. 2015-01-21 20:48:46 -05:00
bunnei
2f3020a102 Mutex: Cleanup and remove redundant code. 2015-01-21 20:48:36 -05:00
bunnei
f09806aed2 Kernel: Renamed some functions for clarity.
- ReleaseNextThread->WakeupNextThread
- ReleaseAllWaitingThreads->WakeupAllWaitingThreads.
2015-01-21 20:48:30 -05:00
bunnei
15b6a4d9ad Kernel: Changed "ShouldWait" to return bool and "Acquire" to return void. 2015-01-21 20:47:49 -05:00
bunnei
c68eb15695 WaitObject: Renamed "Wait" to "ShouldWait", made "ShouldWait" and "Acquire" pure virtual. 2015-01-21 20:47:49 -05:00
bunnei
9412996c8f Kernel: Moved Wait and Acquire to WaitObject, added way to retrieve a WaitObject safely. 2015-01-21 20:47:46 -05:00
bunnei
254e4ebd58 AddressArbiter: Changed to Kernel::Object, big cleanup, removed code that made no sense. 2015-01-21 20:47:45 -05:00
bunnei
e5a9f1c644 Kernel: Get rid of WaitTypes and simplify lots of code, removing hacks. 2015-01-21 20:47:38 -05:00
bunnei
6643673f28 WaitSynchronizationN: Refactor to fix several bugs
- Separate wait checking from waiting the current thread
- Resume thread when wait_all=true only if all objects are available at once
- Set output to correct wait object index when there are duplicate handles
2015-01-21 19:11:47 -05:00
bunnei
aa01c57ae9 Kernel: Separate WaitSynchronization into Wait and Acquire methods. 2015-01-21 19:10:24 -05:00
bunnei
7faf2d8e06 WaitSynchronizationN: Implement return values 2015-01-21 19:09:03 -05:00
bunnei
1f7a04f05a Thread: Keep track of multiple wait objects. 2015-01-21 18:42:04 -05:00
bunnei
c22bac6398 Kernel: Added WaitObject and changed "waitable" objects inherit from it. 2015-01-21 18:41:00 -05:00
bunnei
542b0b0057 Merge pull request #466 from Subv/wake
Thread: Prevent waking a thread multiple times.
2015-01-11 13:39:23 -05:00
Subv
b68d51ed30 Thread: Prevent waking a thread multiple times.
If a thread was woken up by something, cancel the wakeup timeout.
2015-01-11 13:18:52 -05:00
Yuri Kunde Schlesner
8ad41775cc Kernel: Start using boost::intrusive_ptr for lifetime management 2015-01-09 19:43:52 -02:00
Yuri Kunde Schlesner
78ea76e12f Thread: Fix nullptr access in a logging function 2015-01-09 04:02:28 -02:00
Yuri Kunde Schlesner
7f1557fbbd Thread: Rename thread_queue => thread_list 2015-01-09 04:02:17 -02:00
Yuri Kunde Schlesner
9bf8462b96 Thread: Reduce use of Handles and move some funcs to inside the class. 2015-01-09 04:02:15 -02:00
Yuri Kunde Schlesner
ba72208cd4 Kernel: Move Thread's definition to the header file 2015-01-09 03:51:55 -02:00
Yuri Kunde Schlesner
7b3452c730 Move ThreadContext to core/core.h and deal with the fallout 2015-01-09 03:51:55 -02:00
Subv
dfc440785a SVC: Fixed SleepThread.
It will now properly wait the specified number of nanoseconds and then wake up the thread.
2015-01-08 18:39:12 -05:00
Subv
60a373a786 Threads: Use a dummy idle thread when no other are ready.
This thread will not actually execute instructions, it will only advance the timing/events and try to yield immediately to the next ready thread, if there aren't any ready threads then it will be rescheduled and start its job again.
2015-01-07 20:31:31 -05:00
Yuri Kunde Schlesner
122c2bb324 Common: Clean up ThreadQueueList
Replace all the C-style complicated buffer management with a std::deque.
In addition to making the code easier to understand it also adds support
for non-POD IdTypes.

Also clean the rest of the code to follow our code style.
2015-01-07 18:38:25 -02:00
Yuri Kunde Schlesner
7e2903cb74 Kernel: New handle manager
This handle manager more closely mirrors the behaviour of the CTR-OS
one. In addition object ref-counts and support for DuplicateHandle have
been added.

Note that support for DuplicateHandle is still experimental, since parts
of the kernel still use Handles internally, which will likely cause
troubles if two different handles to the same object are used to e.g.
wait on a synchronization primitive.
2014-12-28 11:52:55 -02:00
Yuri Kunde Schlesner
23f2142009 Kernel: Replace GetStaticHandleType by HANDLE_TYPE constants 2014-12-28 11:52:53 -02:00