This change makes for a clearer (less confusing) path of execution in the scheduler, now the code to execute when a thread awakes is closer to the code that puts the thread to sleep (WaitSynch1, WaitSynchN). It also allows us to implement the special wake up behavior of ReplyAndReceive without hacking up WaitObject::WakeupAllWaitingThreads.
If savestates are desired in the future, we can change this implementation to one similar to the CoreTiming event system, where we first register the callback functions at startup and assign their identifiers to the Thread callback variable instead of directly assigning a lambda to the wake up callback variable.
It is unlikely we will ever use this without first doing a Cast to a signed type.
Fixes 9 "unary minus operator applied to unsigned type, result still unsigned" warnings on MSVC2017.3
The loaders will now just create a Kernel::Process, construct it and return it to the caller, which is responsible for setting it as the current process and configuring the global page table.
Don't automatically assume that Thread::Create will only be called when the parent process is currently scheduled. This assumption will be broken when applets or system modules are loaded.
Only use the HLE interface if an HLE applet with the desired id was started.
This commit reorganizes the APT code surrounding parameter creation and delivery to make it easier to support LLE applets in the future.
As future work, the HLE applet interface can be reworked to utilize the same facilities as the LLE interface.
Not all physical regions need to be mapped into the address space of every process, for example, system modules do not have a VRAM mapping.
This fixes a crash when loading applets and system modules.
The current code inserts and deletes elements from the beginning of the audio buffer, which is very inefficient in an std::vector.
Profiling was done using VisualStudio2017's Performance Analyzer in Super Mario 3D Land.
Before this change: AudioInterp::Linear had 14.14% of the runtime (inclusive) and most of that time was spent in std::vector's insert implementation.
After this change: AudioInterp::Linear has 0.36% of the runtime (inclusive)
The loaders now register each loaded ROM with the SelfNCCH factory, which keeps the data around for the duration of the emulation session.
When opening the SelfNCCH archive, the factory queries the current program's programid and uses that as a key to the map that contains the NCCHData structure (RomFS, Icon, Banner, etc).
3dsx files do not have a programid and will use a default of 0 for this value, thus, only 1 3dsx file with RomFS is loadable at the same time.
* loader/ncch: split NCCH parsing into its own file
* loader/ncch: add support for loading update NCCHs from the SD card
* loader/ncch: fix formatting
* file_sys/ncch_container: Return a value for OpenFile
* loader/ncch: cleanup, always instantiate overlay_ncch to base_ncch
* file_sys/ncch_container: better encryption checks, allow non-app NCCHs to load properly and for the existence of NCCH structures to be checked
* file_sys/ncch_container: pass filepath as a const reference
* Services/UDS: Added a function to generate the EAPoL-Start packet body.
* Services/UDS: Added filter for beacons.
* Services/UDS: Lock a mutex when accessing connection_status from both the emulation and network thread.
* Services/UDS: Handle the Association Response frame and respond with the EAPoL-Start frame.
* fixup: make use of current_node, changed received_beacons into a list, mutex and assert corrections
* fixup: fix damn clang-format
As per 3dbrew:
"During Home Menu start-up it uses APT:PrepareToStartNewestHomeMenu. If that doesn't return an error(normally NS returns 0xC8A0CFFC for that), Home Menu starts a hardware reboot with APT:StartNewestHomeMenu etc. "
We need to know what is being run so we can set the APT parameter destination AppId correctly.
Delaying the preparation of the parameter until we know which AppId is running lets us support booting both the Home Menu and normal game Applications.
* WebService: Verify username and token; Log errors in PostJson
* Fixup: added docstrings to the functions
* Webservice: Added Icons to the verification, imrpved error detection in cpr, fixup nits
* fixup: fmt warning
PR #1461 introduced a regression where some games would change configuration
even while in the poorly named "drawing" mode, which broke the heuristic
citra was using to determine when to draw the batch. This change adds
back in a draw call for batching, and also adds in a draw call in
immediate mode each time it adds a triangle.
There is currently no stage tracking, a client is considered "Connected" when it receives the EAPoL Logoff packet from the server, this is not yet implemented.
* added a SidebySide Layout
* Reworked, so both screen have the same height and cleaned up screen translates.
* added the option in the UI, hope this is the right way to do it. formated framebuffer_layout.cpp
* delete the x64 files
* deleted ui_configure_graphics.h
* added Option for the Layout in the xml
* got rid of SIDE_BY_SIDE_ASPECT_RATIO because it was useless. pulled translate into variables
* changed shift variables to u32 and moved them in their respective branch. remove notr="true" for the Screen layout drop down
* reworked intends :). changed function description for SideFrameLayout
* some description reworking
Kernel/HLE: Use a mutex to synchronize access to the HLE kernel state between the cpu thread and any other possible threads that might touch the kernel (network thread, etc).
This mutex is acquired in SVC::CallSVC, ie, as soon as the guest application enters the HLE kernel, and should be acquired by the aforementioned threads before modifying kernel structures.
This is necessary for loading multiple processes at the same time.
The main thread will be automatically scheduled when necessary once the scheduler runs.
