mirrors/rpcs3
1
0
Fork 0
mirror of https://github.com/RPCS3/rpcs3.git synced 2025-04-20 11:36:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

PPU LLVM: Optimize altivec FMA with 0 addend (#8013)

Browse source 
- When VMADDFP and VNMSUBFP are used with a constant addend of 0, they can be simplified into a single floating multiply
This commit is contained in:
Whatcookie 2020-04-12 04:52:21 -04:00 • committed by GitHub
parent 8e61c65c0d
commit 6b0f7a8f55
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 32 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

44
rpcs3/Emu/Cell/PPUTranslator.cpp
View file

@ -904,18 +904,28 @@ void PPUTranslator::VLOGEFP(ppu_opcode_t op)
void PPUTranslator::VMADDFP(ppu_opcode_t op)
{
auto [a, b, c] = get_vrs<f32[4]>(op.va, op.vb, op.vc);
// Optimization: Emit only a floating multiply if the addend is zero
if (auto cv = llvm::dyn_cast<llvm::Constant>(b.value))
{
v128 data = get_const_vector(cv, m_addr, 2000);
if (data == v128{})
{
set_vr(op.vd, a * c);
ppu_log.notice("LLVM: VMADDFP with 0 addend at [0x%08x]", m_addr + (m_reloc ? m_reloc->addr : 0));
return;
}
}
if (m_use_fma)
{
const auto acb = GetVrs(VrType::vf, op.va, op.vc, op.vb);
SetVr(op.vd, m_ir->CreateCall(get_intrinsic<f32[4]>(llvm::Intrinsic::fma), { acb[0], acb[1], acb[2] }));
SetVr(op.vd, m_ir->CreateCall(get_intrinsic<f32[4]>(llvm::Intrinsic::fma), { a.value, c.value, b.value }));
return;
}
// Emulated FMA via double precision
auto a = get_vr<f32[4]>(op.va);
auto b = get_vr<f32[4]>(op.vb);
auto c = get_vr<f32[4]>(op.vc);
const auto xa = m_ir->CreateFPExt(a.value, get_type<f64[4]>());
const auto xb = m_ir->CreateFPExt(b.value, get_type<f64[4]>());
const auto xc = m_ir->CreateFPExt(c.value, get_type<f64[4]>());
@ -1200,19 +1210,29 @@ void PPUTranslator::VMULOUH(ppu_opcode_t op)
void PPUTranslator::VNMSUBFP(ppu_opcode_t op)
{
auto [a, b, c] = get_vrs<f32[4]>(op.va, op.vb, op.vc);
// Optimization: Emit only a floating multiply if the addend is zero
if (auto cv = llvm::dyn_cast<llvm::Constant>(b.value))
{
v128 data = get_const_vector(cv, m_addr, 2004);
if (data == v128{})
{
set_vr(op.vd, -a * c);
ppu_log.notice("LLVM: VNMSUBFP with 0 addend at [0x%08x]", m_addr + (m_reloc ? m_reloc->addr : 0));
return;
}
}
// Differs from the emulated path with regards to negative zero
if (m_use_fma)
{
const auto acb = GetVrs(VrType::vf, op.va, op.vc, op.vb);
SetVr(op.vd, m_ir->CreateFNeg(m_ir->CreateCall(get_intrinsic<f32[4]>(llvm::Intrinsic::fma), { acb[0], acb[1], m_ir->CreateFNeg(acb[2]) })));
SetVr(op.vd, m_ir->CreateFNeg(m_ir->CreateCall(get_intrinsic<f32[4]>(llvm::Intrinsic::fma), { a.value, c.value, m_ir->CreateFNeg(b.value) })));
return;
}
// Emulated FMA via double precision
auto a = get_vr<f32[4]>(op.va);
auto b = get_vr<f32[4]>(op.vb);
auto c = get_vr<f32[4]>(op.vc);
const auto xa = m_ir->CreateFPExt(a.value, get_type<f64[4]>());
const auto xb = m_ir->CreateFPExt(b.value, get_type<f64[4]>());
const auto xc = m_ir->CreateFPExt(c.value, get_type<f64[4]>());