Fundamentally, all this does is enforce the invariant that we always
translate effective addresses based on the current BAT registers and
page table before we do anything else with them.
This change can be logically divided into three parts. The first part is
creating a table to represent the current BAT state, and keeping it up to
date (PowerPC::IBATUpdated, PowerPC::DBATUpdated, etc.). This does
nothing by itself, but it's necessary for the other parts.
The second part (mostly in MMU.cpp) is simply removing all the hardcoded
checks for specific untranslated addresses, and consistently translating
addresses using the current BAT configuration. Very straightforward, but a
lot of code changes because we hardcoded assumptions all over the place.
The third part (mostly in Memmap.cpp) is making the fastmem arena reflect
the current BAT configuration. We do this by redoing the mapping (calling
memmap()) based on the BAT table whenever it changes.
One additional minor change is that translation can fail in two ways:
either the segment is a direct store segment, or page table lookup failed.
The difference doesn't usually matter, but the difference affects cache
instructions, like dcbz.
Init cannot be called more than once because it registers the
CoreTiming callbacks, that trips the assertions and will cause
anyone with PanicAlerts disabled to crash.
CoreTiming gets restored before ExpansionInterface so CoreTiming
events need to already be registered before the save state loading
begins. This means that the callbacks must be registered
unconditionally instead of on-demand.
Replace adhoc linked list with a priority heap. Performance
characteristics are mostly the same, but is more cache friendly.
[Priority Queues have O(log n) push/pop compared to the linked
list's O(n) push/O(1) pop but the queue is not big enough for
that to matter, so linear is faster over linked. Very slight gains
when framelimit is unlimited (Wind Waker), 1900% -> 1950%]
When n was a multiple of 4, the old implementation would overwrite
the following register with 0.
This was causing Not64 to crash.
Thanks to Extrems for spotting this.
I'm not sure this is the correct fix, but it looks like OSREPORT output
is Shift-JIS, so we need to convert it to UTF-8. Most characters work
fine without and with this conversion, but Japanese text completely
fails and results in outputting invalid UTF-8 (which gets shown as �).
The gather pipe optimization postpones checking the FIFO until the end
of the current block (or 32 bytes have been written). This is usually
safe, but is not correct across EIEIO instructions.
This is inferred from a block in NBA2K11 which synchronizes the FIFO
by writing a byte to it, executing eieio, and checking if PI_FIFO_WPTR
has changed. This is not currently an issue, but will become an issue
if the gather pipe optimization is applied to more stores.
Let's stop pretending that we support Triforce emulation.
Keeping this code around just in case someone will make
major improvements in the future isn't really worth it.
I'm keeping the Triforce game INIs so users will know that
the compatibility rating for Triforce games is 1 star (broken).
These operations should always take the same amount of time,
not the same amount of ticks. The number of ticks per second
is different for GameCube and Wii.
Replaces old and simple usages of std::atomic<bool> with Common::Flag
(which was introduced after the initial usage), so it's clear that
the variable is a flag and because Common::Flag is well tested.
This also replaces the ready logic in WiimoteReal with Common::Event
since it was basically just unnecessarily reimplementing Common::Event.
