It turns out that the offsets of the data in the return buffer matter: the
offset 0x340 is hardcoded into the code that uses the result.
With that fixed, get rid of the completely wrong 0x80000000 hack.
psq_st performs one store, and psq_ld one load, from the perspective of the
MMU; getting this wrong leads to potentially incorrect behavior (incorrect page
faults, weirdness with the gather pipe, etc.). Fix this, and stop masking
the address when checking for gather pipe writes.
Also a bunch of cleanup.
This came up from the discussion we were having prior about dumping half of a kilobyte of VFP registers to the stack is insanity.
This was due to me basically copying exactly what I did on ARMv7's paired loadstores, where the impact is less since we only use the bottom 64bits of
the VFP registers.
So I decided to think about how to improve upon this since I got called out on my terrible code.
The solution I have come up with is instead of jumping to the common ASM routine and having that check if it needs to take the fastmem or slowmem
routes, just inline the check in to the JIT block and jump to either a fastmem or slowmem handler.
Fairly simple, and this allows us to only flush the registers that are required when doing so. Should give a reasonable increase in performance for
games that use the slowmem path quite a lot.
mathieui
skidau
Lioncash
Stevoisiak
Markus Wick
Rachel Bryk
magumagu
Dolphin Bot
JosJuice
Ryan Houdek
degasus