ShuriZma
/
pcsx2
mirror of https://github.com/PCSX2/pcsx2.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

gsdx sw x64: prefer faster 32 bits operation when possible

This commit is contained in:
Gregory Hainaut 2016-11-18 16:48:25 +01:00
parent a281bda9a6
commit 141c9e9c86
1 changed files with 25 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

47
plugins/GSdx/GSDrawScanlineCodeGenerator.x64.avx.cpp
View File

@ -72,8 +72,10 @@ void GSDrawScanlineCodeGenerator::Generate()
mov(ptr[rsp - 1 * 8], rbx);
mov(ptr[rsp - 2 * 8], r12);
mov(ptr[rsp - 3 * 8], r13);
mov(ptr[rsp - 4 * 8], r14);
mov(ptr[rsp - 5 * 8], r15);
if(need_clut)
mov(ptr[rsp - 4 * 8], r14);
if(need_tex)
mov(ptr[rsp - 5 * 8], r15);
#endif
mov(r10, (size_t)&m_test[0]);
@ -81,9 +83,10 @@ void GSDrawScanlineCodeGenerator::Generate()
mov(_m_local__gd, ptr[_m_local + offsetof(GSScanlineLocalData, gd)]);
mov(_m_local__gd__vm, ptr[_m_local__gd + offsetof(GSScanlineGlobalData, vm)]);
// FIXME: those 2 load could be optimized when no texture
mov(_m_local__gd__clut, ptr[_m_local__gd + offsetof(GSScanlineGlobalData, clut)]);
mov(_m_local__gd__tex, ptr[_m_local__gd + offsetof(GSScanlineGlobalData, tex)]);
if(need_clut)
mov(_m_local__gd__clut, ptr[_m_local__gd + offsetof(GSScanlineGlobalData, clut)]);
if(need_tex)
mov(_m_local__gd__tex, ptr[_m_local__gd + offsetof(GSScanlineGlobalData, tex)]);
Init();
@ -252,8 +255,10 @@ L("exit");
mov(rbx, ptr[rsp - 1 * 8]);
mov(r12, ptr[rsp - 2 * 8]);
mov(r13, ptr[rsp - 3 * 8]);
mov(r14, ptr[rsp - 4 * 8]);
mov(r15, ptr[rsp - 5 * 8]);
if(need_clut)
mov(r14, ptr[rsp - 4 * 8]);
if(need_tex)
mov(r15, ptr[rsp - 5 * 8]);
pop(rbp);
#endif
@ -266,12 +271,12 @@ void GSDrawScanlineCodeGenerator::Init()
{
// int skip = left & 3;
mov(rbx, a1);
and(a1, 3);
mov(ebx, a1.cvt32());
and(a1.cvt32(), 3);
// left -= skip;
sub(rbx, a1);
sub(ebx, a1.cvt32());
// int steps = pixels + skip - 4;
@ -279,7 +284,7 @@ void GSDrawScanlineCodeGenerator::Init()
// GSVector4i test = m_test[skip] | m_test[7 + (steps & (steps >> 31))];
shl(a1, 4); // * sizeof(m_test[0])
shl(a1.cvt32(), 4); // * sizeof(m_test[0])
vmovdqa(_test, ptr[a1 + r10]);
@ -292,8 +297,8 @@ void GSDrawScanlineCodeGenerator::Init()
}
else
{
mov(rbx, a1); // left
xor(a1, a1); // skip
mov(ebx, a1.cvt32()); // left
xor(a1.cvt32(), a1.cvt32()); // skip
lea(a0, ptr[a0 - 4]); // steps
}
@ -591,10 +596,9 @@ void GSDrawScanlineCodeGenerator::TestZ(const Xmm& temp1, const Xmm& temp2)
// int za = fza_base.y + fza_offset->y;
movsxd(rbp, dword[t1 + 4]);
movsxd(rax, dword[t0 + 4]);
add(rbp, rax);
and(rbp, HALF_VM_SIZE - 1);
mov(ebp, dword[t1 + 4]);
add(ebp, dword[t0 + 4]);
and(ebp, HALF_VM_SIZE - 1);
// GSVector4i zs = zi;
@ -1413,7 +1417,6 @@ void GSDrawScanlineCodeGenerator::ReadFrame()
mov(ebx, dword[t1]);
add(ebx, dword[t0]);
and(ebx, HALF_VM_SIZE - 1);
movsxd(rbx, ebx); // FIXME useful ?
if(!m_sel.rfb)
{
@ -1776,9 +1779,9 @@ void GSDrawScanlineCodeGenerator::WriteFrame()
// y = (top & 3) << 5
mov(rax, a1);
and(rax, 3);
shl(rax, 5);
mov(eax, a1.cvt32());
and(eax, 3);
shl(eax, 5);
// rb = rb.add16(m_global.dimx[0 + y]);
// ga = ga.add16(m_global.dimx[1 + y]);
@ -1977,7 +1980,7 @@ void GSDrawScanlineCodeGenerator::ReadTexel(const Xmm& dst, const Xmm& addr, uin
//else vpextrd(eax, addr, i);
vpextrd(eax, addr, i);
if(m_sel.tlu) movzx(rax, byte[_m_local__gd__tex + rax]);
if(m_sel.tlu) movzx(eax, byte[_m_local__gd__tex + rax]);
//if(i == 0) vmovd(dst, src);
//else vpinsrd(dst, src, i);