The free-block tracker search in BaseHeap::AllocRange treats
high_page_number as inclusive, shifting allocator-returned addresses
by one stride relative to the old loop-based search. Some titles
encode allocator-returned addresses in PPC code and break when that
layout shifts (Far Cry 3, Far Cry 4, Watchdogs).
In addition, reapply xe::align on the high side of AllocRange
(essentially reverting c28019e33). Without the round-up, a caller
passing a min/max window exactly the size of its request loses a
stride at the top and fails the early page_count size check.
BaseHeap::AllocRange was rounding the inclusive high_address UP via
xe::align(), which could push the search range one alignment stride
past the requested end. The new free-block-tracker top-down search
treats high_page_number as an inclusive max (usable_end =
high_page_number + 1), so the round-up caused it to return a base
page one stride beyond the caller's bound — most visibly when
PhysicalHeap::Alloc passed parent_heap_end = GetPhysicalAddress(...)
just below a page boundary, producing a translated address one page
past the child heap and tripping "passed out of range address range"
in BaseHeap::AllocFixed.
Drop the xe::align on the high side so high_address stays a true
inclusive bound. The low_address round-up is still correct since
allocations must START at or above the aligned low.
Replace linear page_table_ scans in AllocRange with a std::map-based
free block index that tracks contiguous free regions.
Insertions now coalesce with adjacent blocks on release and AllocFixed
uses the targeted tracker for pure reserves and falls back to a full rebuild
for mixed-state commits.
Also fixing PhysicalHeap leaking parent memory on child allocation failure,
Reset() not restoring unreserved_page_count_ and some incorrect method names
in PhysicalHeap error messages
The threading test is verifying that the timer fires, not that it fires
within a specific timeout, so increasing to 100 to keep it from failing
randomly. Also removing xenia-cpu-ppc-tests from list of default targets
as they take forever to run, they can still be run run using xb test
with --target xenia-cpu-ppc-tests explicitly.
- Mainly use `assert`s, since failure is very rare
- Forward failure of `CreateSemaphore` to guests because it is more easy
to trigger with invalid initial parameters.
Timing dependencies in this tests were causing spurious test failures:
- Create and Run Thread
- Test Thread QueueUserCallback
They have been largely replaced by spin waits.
Rather than using `n * 1024 * 1024`, this adds a convenient `_MiB`/`_KiB` user-literal to the new `literals.h` header to concisely describe units of memory in a much more readable way. Any other useful literals can be added to this header. These literals exist in the `xe::literals` namespace so they are opt-in, similar to `std::chrono` literals, and require a `using namespace xe::literals` statement to utilize it within the current scope.
I've done a pass through the codebase to replace trivial instances of `1024 * 1024 * ...` expressions being used but avoided anything that added additional casting complexity from `size_t` to `uint32_t` and such to keep this commit concise.
shm_unlink(name) is the proper way to close a shared memory in linux.
Prior to this, xenia was creating and not cleaning up shared memory handle
which would accumulate in /dev/shm. shm_unlink is the proper way of doing
this.
Add filename to CloseFileMappingHandle signature.
Add simple test to open and close.
CreateFileMappingHandle now takes shared memory name without a prefix.
The doc of shm_open recommends not using slashes and prefixing with "/".
The prefixing has been moved to the os implementation layer.
Invocations of CreateFileMappingHandle were all using "Local\\" so these
prefixes were removed.
Add suspend count to thread implementation.
Increment suspend count on suspend and decrement on resume.
Wait on suspend count to be decremented to 0.
Return suspend count on suspend and on resume before incr/decr.
Fix naming of resume suspend count to make clear that suspend count is
before incr/decr.
Add test.
Add thread local bool for alertable state.
Use real-time event interrupt to run callback.
Fix sleep duration from miliseconds (microseconds / 1000) to seconds in sleep
command.
Add note for future implementation.
Ignore real-time event 37 in .gdbinit which is used to signal callback.
Test AlertableSleep
Test Thread QueueUserCallback.
TODO: Test alerted wait result when using IO functions.
Use real-time event interrupt to communicate suspend in timely manner.
Use conditional_variable to implement suspend wait and resume trigger.
Ignore real-time event 36 in .gdbinit which is used to signal suspend.
Test suspending threads.
