This should be comparing against the queried process' vma_map, not the
current process'. The only reason this hasn't become an issue yet is we
currently only handle one process being active at any time.
Now that the changes clarifying the address spaces has been merged, we
can wrap the checks that the kernel performs when mapping shared memory
(and other forms of memory) into its own helper function and then use
those within MapSharedMemory and UnmapSharedMemory to complete the
sanitizing checks that are supposed to be done.
So, one thing that's puzzled me is why the kernel seemed to *not* use
the direct code address ranges in some cases for some service functions.
For example, in svcMapMemory, the full address space width is compared
against for validity, but for svcMapSharedMemory, it compares against
0xFFE00000, 0xFF8000000, and 0x7FF8000000 as upper bounds, and uses
either 0x200000 or 0x8000000 as the lower-bounds as the beginning of the
compared range. Coincidentally, these exact same values are also used in
svcGetInfo, and also when initializing the user address space, so this
is actually retrieving the ASLR extents, not the extents of the address
space in general.
This should help diagnose crashes easier and prevent many users thinking that a game is still running when in fact it's just an audio thread still running(this is typically not killed when svcBreak is hit since the game expects us to do this)
A fairly basic service function, which only appears to currently support
retrieving the process state. This also alters the ProcessStatus enum to
contain all of the values that a kernel process seems to be able of
reporting with regards to state.
These only exist to ferry data into a Process instance and end up going
out of scope quite early. Because of this, we can just make it a plain
struct for holding things and just std::move it into the relevant
function. There's no need to make this inherit from the kernel's Object
type.
Regular value initialization is adequate here for zeroing out data. It
also has the benefit of not invoking undefined behavior if a non-trivial
type is ever added to the struct for whatever reason.
This adds the missing address range checking that the service functions
do before attempting to map or unmap memory. Given that both service
functions perform the same set of checks in the same order, we can wrap
these into a function and just call it from both functions, which
deduplicates a little bit of code.
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).
When loading NROs, svcBreak is called to signal to the debugger that a new "module" is loaded. As no debugger is technically attached we shouldn't be killing the programs execution.
This was the result of a typo accidentally introduced in
e51d715700. This restores the previous
correct behavior.
The behavior with the reference was incorrect and would cause some games
to fail to boot.
Conceptually, it doesn't make sense for a thread to be able to persist
the lifetime of a scheduler. A scheduler should be taking care of the
threads; the threads should not be taking care of the scheduler.
If the threads outlive the scheduler (or we simply don't actually
terminate/shutdown the threads), then it should be considered a bug
that we need to fix.
Attributing this to balika011, as they opened #1317 to attempt to fix
this in a similar way, but my refactoring of the kernel code caused
quite a few conflicts.
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.
Now that we have all of the rearranging and proper structure sizes in
place, it's fairly trivial to implement svcGetThreadContext(). In the
64-bit case we can more or less just write out the context as is, minus
some minor value sanitizing. In the 32-bit case we'll need to clear out
the registers that wouldn't normally be accessible from a 32-bit
AArch32 exectuable (or process).
This will be necessary for the implementation of svcGetThreadContext(),
as the kernel checks whether or not the process that owns the thread
that has it context being retrieved is a 64-bit or 32-bit process.
If the process is 32-bit, then the upper 15 general-purpose registers
and upper 16 vector registers are cleared to zero (as AArch32 only has
15 GPRs and 16 128-bit vector registers. not 31 general-purpose
registers and 32 128-bit vector registers like AArch64).
Makes the public interface consistent in terms of how accesses are done
on a process object. It also makes it slightly nicer to reason about the
logic of the process class, as we don't want to expose everything to
external code.
boost::static_pointer_cast for boost::intrusive_ptr (what SharedPtr is),
takes its parameter by const reference. Given that, it means that this
std::move doesn't actually do anything other than obscure what the
function's actual behavior is, so we can remove this. To clarify, this
would only do something if the parameter was either taking its argument
by value, by non-const ref, or by rvalue-reference.
The locations of these can actually vary depending on the address space
layout, so we shouldn't be using these when determining where to map
memory or be using them as offsets for calculations. This keeps all the
memory ranges flexible and malleable based off of the virtual memory
manager instance state.
Previously, these were reporting hardcoded values, but given the regions
can change depending on the requested address spaces, these need to
report the values that the memory manager contains.
Rather than hard-code the address range to be 36-bit, we can derive the
parameters from supplied NPDM metadata if the supplied exectuable
supports it. This is the bare minimum necessary for this to be possible.
The following commits will rework the memory code further to adjust to
this.
The owning process of a thread is required to exist before the thread,
so we can enforce this API-wise by using a reference. We can also avoid
the reliance on the system instance by using that parameter to access
the page table that needs to be set.
This can just be a regular function, getting rid of the need to also
explicitly undef the define at the end of the file. Given FuncReturn()
was already converted into a function, it's #undef can also be removed.
This modifies the CPU interface to more accurately match an
AArch64-supporting CPU as opposed to an ARM11 one. Two of the methods
don't even make sense to keep around for this interface, as Adv Simd is
used, rather than the VFP in the primary execution state. This is
essentially a modernization change that should have occurred from the
get-go.
The kernel does the equivalent of the following check before proceeding:
if (address + 0x8000000000 < 0x7FFFE00000) {
return ERR_INVALID_MEMORY_STATE;
}
which is essentially what our IsKernelVirtualAddress() function does. So
we should also be checking for this.
The kernel also checks if the given input addresses are 4-byte aligned,
however our Mutex::TryAcquire() and Mutex::Release() functions already
handle this, so we don't need to add code for this case.
The kernel caps the size limit of shared memory to 8589930496 bytes (or
(1GB - 512 bytes) * 8), so approximately 8GB, where every GB has a 512
byte sector taken off of it.
It also ensures the shared memory is created with either read or
read/write permissions for both permission types passed in, allowing the
remote permissions to also be set as "don't care".
Part of the checking done by the kernel is to check if the given
address and size are 4KB aligned, as well as checking if the size isn't
zero. It also only allows mapping shared memory as readable or
read/write, but nothing else, and so we shouldn't allow mapping as
anything else either.
Previously, these were sitting outside of the Kernel namespace, which
doesn't really make sense, given they're related to the Thread class
which is within the Kernel namespace.
While unlikely, it does avoid constructing a std::string and
unnecessarily calling into the memory code if a game or executable
decides to be really silly about their logging.
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.
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.
The follow-up to e2457418da, which
replaces most of the includes in the core header with forward declarations.
This makes it so that if any of the headers the core header was
previously including change, then no one will need to rebuild the bulk
of the core, due to core.h being quite a prevalent inclusion.
This should make turnaround for changes much faster for developers.
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.
We can make this error code an alias of the resource limit exceeded
error code, allowing us to get rid of the lingering 3DS error code of
the same type.
We already have the variable itself set up to perform this task, so we
can just return its value from the currently executing process instead
of always stubbing it to zero.
Allows querying the inverse of IsDomain() to make things more readable.
This will likely also be usable in the event of implementing
ConvertDomainToSession().
Despite being covered by a global mutex, we should still ensure that the
class handles its reference counts properly. This avoids potential
shenanigans when it comes to data races.
Given this is the root object that drives quite a bit of the kernel
object hierarchy, ensuring we always have the correct behavior (and no
races) is a good thing.
The current core may have nothing to do with the core where the new thread was scheduled to run. In case it's the same core, then the following PrepareReshedule call will take care of that.
WakeAfterDelay might be called from any host thread, so err on the side of caution and use the thread-safe CoreTiming::ScheduleEventThreadsafe.
Note that CoreTiming is still far from thread-safe, there may be more things we have to work on for it to be up to par with what we want.
Exit from AddMutexWaiter early if the thread is already waiting for a mutex owned by the owner thread.
This accounts for the possibility of a thread that is waiting on a condition variable being awakened twice in a row.
Also added more validation asserts.
This should fix one of the random crashes in Breath Of The Wild.
These members don't need to be entirely exposed, we can instead expose
an API to operate on them without directly needing to mutate them
We can also guard against overflow/API misuse this way as well, given
active_sessions is an unsigned value.
This amends cases where crashes can occur that were missed due to the
odd way the previous code was set up (using 3DS memory regions that
don't exist).
Using member variables for referencing the segments array increases the
size of the class in memory for little benefit. The same behavior can be
achieved through the use of accessors that just return the relevant
segment.
Avoids using a u32 to compare against a range of size_t, which can be a
source of warnings. While we're at it, compress a std::tie into a
structured binding.
General moving to keep kernel object types separate from the direct
kernel code. Also essentially a preliminary cleanup before eliminating
global kernel state in the kernel code.
This introduces a slightly more generic variant of WriteBuffer().
Notably, this variant doesn't constrain the arguments to only accepting
std::vector instances. It accepts whatever adheres to the
ContiguousContainer concept in the C++ standard library.
This essentially means, std::array, std::string, and std::vector can be
used directly with this interface. The interface no longer forces you to
solely use containers that dynamically allocate.
To ensure our overloads play nice with one another, we only enable the
container-based WriteBuffer if the argument is not a pointer, otherwise
we fall back to the pointer-based one.
The reason this would never be true is that ideal_processor is a u8 and
THREADPROCESSORID_DEFAULT is an s32. In this case, it boils down to how
arithmetic conversions are performed before performing the comparison.
If an unsigned value has a lesser conversion rank (aka smaller size)
than the signed type being compared, then the unsigned value is promoted
to the signed value (i.e. u8 -> s32 happens before the comparison). No
sign-extension occurs here either.
An alternative phrasing:
Say we have a variable named core and it's given a value of -2.
u8 core = -2;
This becomes 254 due to the lack of sign. During integral promotion to
the signed type, this still remains as 254, and therefore the condition
will always be true, because no matter what value the u8 is given it
will never be -2 in terms of 32 bits.
Now, if one type was a s32 and one was a u32, this would be entirely
different, since they have the same bit width (and the signed type would
be converted to unsigned instead of the other way around) but would
still have its representation preserved in terms of bits, allowing the
comparison to be false in some cases, as opposed to being true all the
time.
---
We also get rid of two signed/unsigned comparison warnings while we're
at it.
Previously, the buffer_index parameter was unused, causing all writes to
use the buffer index of zero, which is not necessarily what is wanted
all the time.
Thankfully, all current usages don't use a buffer index other than zero,
so this just prevents a bug before it has a chance to spring.
This would result in a lot of allocations and related object
construction, just to toss it all away immediately after the call.
These are definitely not intentional, and it was intended that all of
these should have been accessing the static function GetInstance()
through the name itself, not constructed instances.
This situation may happen like so:
Thread 1 with low priority calls WaitProcessWideKey with timeout.
Thread 2 with high priority calls WaitProcessWideKey without timeout.
Thread 3 calls SignalProcessWideKey
- Thread 2 acquires the lock and awakens.
- Thread 1 can't acquire the lock and is put to sleep with the lock owner being Thread 2.
Thread 1's timeout expires, with the lock owner still being set to Thread 2.
This makes the formatting expectations more obvious (e.g. any zero padding specified
is padding that's entirely dedicated to the value being printed, not any pretty-printing
that also gets tacked on).
* GetSharedFontInOrderOfPriority
* Update pl_u.cpp
* Ability to use ReadBuffer and WriteBuffer with different buffer indexes, fixed up GetSharedFontInOrderOfPriority
* switched to NGLOG
* Update pl_u.cpp
* Update pl_u.cpp
* language_code is actually language code and not index
* u32->u64
* final cleanups
Verified with a hwtest and implemented based on reverse engineering.
Thread A's priority will get bumped to the highest priority among all the threads that are waiting for a mutex that A holds.
Once A releases the mutex and ownership is transferred to B, A's priority will return to normal and B's priority will be bumped.
Switch mutexes are no longer kernel objects, they are managed in userland and only use the kernel to handle the contention case.
Mutex addresses store a special flag value (0x40000000) to notify the guest code that there are still some threads waiting for the mutex to be released. This flag is updated when a thread calls ArbitrateUnlock.
TODO:
* Fix svcWaitProcessWideKey
* Fix svcSignalProcessWideKey
* Remove the Mutex class.
* Updated ACC with more service names
* Updated SVC with more service names
* Updated set with more service names
* Updated sockets with more service names
* Updated SPL with more service names
* Updated time with more service names
* Updated vi with more service names
Ported from citra PR #3091
The delay specified here is from a Nintendo 3DS, and should be measured in a Nintendo Switch.
This change is enough to prevent Puyo Puyo Tetris's main thread starvation.
Makes the codebase a little more consistent with regards to available documentation. Also amends the duplicate case where there was a similar entry at 0x72 named ConnectToPort.
added more svcGetInfo pairs for 3.0.0+ support, Changed HEAP_SIZE and TLS_AREA_VADDR. changed mem usage & heap usage stub added, ISelfController, IApplication function stubs. Added SetThreadCoreMask
Also properly keep track of data in guest memory, this fixes managing the semaphore from userland.
It was found that Semaphores are actually Condition Variables, with Release(1) and Release(-1) being equivalent to notify_one and notify_all. We should change the name of the class to reflect this.
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.
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.
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.
Copy the IPC command buffer to/from the request context before/after the
handler is invoked. This is part of a move away from using global data
for handling IPC requests.
The old "Interface" class had a few problems such as using free
functions (Which didn't allow you to write the service handler as if it
were a regular class.) which weren't very extensible. (Only received one
parameter with a pointer to the Interface object.)
The new ServiceFramework aims to solve these problems by working with
member functions and passing a generic context struct as parameter. This
struct can be extended in the future without having to update all
existing service implementations.
This allows attaching a HLE handle to a ServerPort at any point after it
is created, allowing port/session creation to be generic between HLE and
regular services.
This replaces the hardcoded VRAM/DSP mappings with ones made based on
the ExHeader ARM11 Kernel caps list. While this has no visible effect
for most applications (since they use a standard set of mappings) it
does improve support for system modules and n3DS exclusives.
Corrects a few issues with regards to Doxygen documentation, for example:
- Incorrect parameter referencing.
- Missing @param tags.
- Typos in @param tags.
and a few minor other issues.
After hwtesting and reverse engineering the kernel, it was found that the CTROS scheduler performs no priority boosting for threads like this, although some other forms of scheduling priority-starved threads might take place.
For example, it was found that hardware interrupts might cause low-priority threads to run if the CPU is preempted in the middle of an SVC handler that deschedules the current (high priority) thread before scheduling it again.
This fixes a potential bug where threads would not get removed from said list if they awoke after waiting with WaitSynchronizationN with wait_all = false
This commit removes the overly general THREADSTATUS_WAIT_SYNCH and replaces it with two more granular statuses:
THREADSTATUS_WAIT_SYNCH_ANY when a thread waits on objects via WaitSynchronization1 or WaitSynchronizationN with wait_all = false.
THREADSTATUS_WAIT_SYNCH_ALL when a thread waits on objects via WaitSynchronizationN with wait_all = true.
The implementation is based on reverse engineering of the 3DS's kernel.
A mutex holder's priority will be temporarily boosted to the best priority among any threads that want to acquire any of its held mutexes.
When the holder releases the mutex, it's priority will be boosted to the best priority among the threads that want to acquire any of its remaining held mutexes.
Define a variable with the value of the sync timeout error code.
Use a boost::flat_map instead of an unordered_map to hold the equivalence of objects and wait indices in a WaitSynchN call.
Threads will now be awakened when the objects they're waiting on are signaled, instead of repeating the WaitSynchronization call every now and then.
The scheduler is now called once after every SVC call, and once after a thread is awakened from sleep by its timeout callback.
This new implementation is based off reverse-engineering of the real kernel.
See https://gist.github.com/Subv/02f29bd9f1e5deb7aceea1e8f019c8f4 for a more detailed description of how the real kernel handles rescheduling.
Sessions and Ports are now detached from each other.
HLE services are handled by means of a SessionRequestHandler class, Interface now inherits from this class.
The File and Directory classes are no longer kernel objects, but SessionRequestHandlers instead, bound to a ServerSession when requested.
File::OpenLinkFile now creates a new session pair and binds the File instance to it.
All handles obtained via srv::GetServiceHandle or svcConnectToPort are references to ClientSessions.
Service modules will wait on the counterpart of those ClientSessions (Called ServerSessions) using svcReplyAndReceive or svcWaitSynchronization[1|N], and will be awoken when a SyncRequest is performed.
HLE Interfaces are now ClientPorts which override the HandleSyncRequest virtual member function to perform command handling immediately.
Applications can request the kernel to allocate a piece of the linear heap for them when creating a shared memory object.
Shared memory areas are now properly mapped into the target processes when calling svcMapMemoryBlock.
Removed the APT Shared Font hack as it is no longer needed.
Each thread gets a 0x200-byte area from the 0x1000-sized page, when all 8 thread slots in a single page are used up, the kernel allocates a new page to hold another 8 entries.
This is consistent with what the real kernel does.
This commit:
* Adds a new subproject, audio_core.
* Defines structures that exist in DSP shared memory.
* Hooks up various other parts of the emulator into audio core.
This sets the foundation for a later HLE DSP implementation.
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.
This commit fixes several kernel object leaks. The most severe of them
was threads not being removed from the private handle table used for
CoreTiming events. This resulted in Threads never being released, which
in turn held references to Process, causing CodeSets to never be freed
when loading other applications.
The code now properly configures the process image to match the loaded
binary segments (code, rodata, data) instead of just blindly allocating
a large chunk of dummy memory.
This enables more dynamic management of the process address space,
compared to just directly configuring the page table for major areas.
This will serve as the foundation upon which the rest of the Kernel
memory management functions will be built.
Implemented svcs GetResourceLimit, GetResourceLimitCurrentValues and GetResourceLimitLimitValues.
Note that the resource limits do not currently keep track of used objects, since we have no way to distinguish between an object created by the application, and an object created by some HLE module once we're inside Kernel::T::Create.
memory.cpp/h contains definitions related to acessing memory and
configuring the address space
mem_map.cpp/h contains higher-level definitions related to configuring
the address space accoording to the kernel and allocating memory.
This works around crashes related to GSP/HID/etc. shared memory blocks
having garbage values. The proper fix requires proper management of
mapped memory blocks in the process.
When the macro was introduced in 326ec51261
it wasn't noticed that it conflicted in name with a heavily used macro
inside of dyncom. This causes some compiler warnings. Since it's only
lightly used, it was opted to simply remove the new macro.
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.
* 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.
