Browse Source

[dynarmic] disable xmm spills with Windows ABI

Signed-off-by: lizzie <lizzie@eden-emu.dev>
pull/203/head
lizzie 7 months ago
committed by crueter
parent
commit
790671f017
  1. 13
      src/dynarmic/src/dynarmic/backend/x64/reg_alloc.cpp

13
src/dynarmic/src/dynarmic/backend/x64/reg_alloc.cpp

@ -567,22 +567,23 @@ void RegAlloc::SpillRegister(HostLoc loc) noexcept {
}
HostLoc RegAlloc::FindFreeSpill(bool is_xmm) const noexcept {
#ifdef _WIN32
// TODO(lizzie): Ok, Windows hates XMM spills, this means less perf for windows
// but it's fine anyways. We can find other ways to cheat it later - but which?!?!
// we should NOT save xmm each block entering... MAYBE xbyak has a bug on start/end?
// TODO(lizzie): This needs to be investigated further later.
#else
// Do not spill XMM into other XMM silly
if (!is_xmm) {
// TODO(lizzie): Using lower (xmm0 and such) registers results in issues/crashes - INVESTIGATE WHY
// Intel recommends to spill GPR onto XMM registers IF POSSIBLE
// TODO(lizzie): Issues on DBZ, theory: Scratch XMM not properly restored after a function call?
// Must sync with ABI registers (except XMM0, XMM1 and XMM2)
#ifdef _WIN32
for (size_t i = size_t(HostLoc::XMM5); i >= size_t(HostLoc::XMM3); --i)
if (const auto loc = HostLoc(i); LocInfo(loc).IsEmpty())
return loc;
#else
for (size_t i = size_t(HostLoc::XMM15); i >= size_t(HostLoc::XMM3); --i)
if (const auto loc = HostLoc(i); LocInfo(loc).IsEmpty())
return loc;
#endif
}
#endif
// Otherwise go to stack spilling
for (size_t i = size_t(HostLoc::FirstSpill); i < hostloc_info.size(); ++i)
if (const auto loc = HostLoc(i); LocInfo(loc).IsEmpty())

Loading…
Cancel
Save