Minor iCore.c changes. A few if/else statements are now case statements, a few compiler warnings are taken care of, a few variables are no longer assigned inside if statements, etc...

git-svn-id: http://pcsx2-playground.googlecode.com/svn/trunk@337 a6443dda-0b58-4228-96e9-037be469359c
This commit is contained in:
arcum42 2008-11-18 11:19:35 +00:00 committed by Gregory Hainaut
parent d24add597e
commit 098e6b509b
2 changed files with 203 additions and 145 deletions

View File

@ -108,17 +108,28 @@ void _initXMMregs() {
__forceinline void* _XMMGetAddr(int type, int reg, VURegs *VU) __forceinline void* _XMMGetAddr(int type, int reg, VURegs *VU)
{ {
if (type == XMMTYPE_VFREG ) return (void*)VU_VFx_ADDR(reg); switch (type) {
else if (type == XMMTYPE_ACC ) return (void*)VU_ACCx_ADDR; case XMMTYPE_VFREG:
else if (type == XMMTYPE_GPRREG) { return (void*)VU_VFx_ADDR(reg);
if( reg < 32 )
assert( !(g_cpuHasConstReg & (1<<reg)) || (g_cpuFlushedConstReg & (1<<reg)) ); case XMMTYPE_ACC:
return &cpuRegs.GPR.r[reg].UL[0]; return (void*)VU_ACCx_ADDR;
case XMMTYPE_GPRREG:
if( reg < 32 )
assert( !(g_cpuHasConstReg & (1<<reg)) || (g_cpuFlushedConstReg & (1<<reg)) );
return &cpuRegs.GPR.r[reg].UL[0];
case XMMTYPE_FPREG:
return &fpuRegs.fpr[reg];
case XMMTYPE_FPACC:
return &fpuRegs.ACC.f;
default:
assert(0);
} }
else if (type == XMMTYPE_FPREG ) return &fpuRegs.fpr[reg];
else if (type == XMMTYPE_FPACC ) return &fpuRegs.ACC.f;
assert(0);
return NULL; return NULL;
} }
@ -205,8 +216,10 @@ int _allocVFtoXMMreg(VURegs *VU, int xmmreg, int vfreg, int mode) {
int readfromreg = -1; int readfromreg = -1;
for (i=0; i<XMMREGS; i++) { for (i=0; i<XMMREGS; i++) {
if (xmmregs[i].inuse == 0 || xmmregs[i].type != XMMTYPE_VFREG || xmmregs[i].reg != vfreg || xmmregs[i].VU != XMM_CONV_VU(VU) ) continue; if ((xmmregs[i].inuse == 0) || (xmmregs[i].type != XMMTYPE_VFREG) ||
(xmmregs[i].reg != vfreg) || (xmmregs[i].VU != XMM_CONV_VU(VU)))
continue;
if( xmmreg >= 0 ) { if( xmmreg >= 0 ) {
// requested specific reg, so return that instead // requested specific reg, so return that instead
if( i != xmmreg ) { if( i != xmmreg ) {
@ -231,12 +244,10 @@ int _allocVFtoXMMreg(VURegs *VU, int xmmreg, int vfreg, int mode) {
return i; return i;
} }
if (xmmreg == -1) { if (xmmreg == -1)
xmmreg = _getFreeXMMreg(); xmmreg = _getFreeXMMreg();
} else
else {
_freeXMMreg(xmmreg); _freeXMMreg(xmmreg);
}
g_xmmtypes[xmmreg] = XMMT_FPS; g_xmmtypes[xmmreg] = XMMT_FPS;
xmmregs[xmmreg].inuse = 1; xmmregs[xmmreg].inuse = 1;
@ -259,7 +270,7 @@ int _checkXMMreg(int type, int reg, int mode)
int i; int i;
for (i=0; i<XMMREGS; i++) { for (i=0; i<XMMREGS; i++) {
if (xmmregs[i].inuse && xmmregs[i].type == (type&0xff) && xmmregs[i].reg == reg ) { if (xmmregs[i].inuse && (xmmregs[i].type == (type&0xff)) && (xmmregs[i].reg == reg)) {
if ( !(xmmregs[i].mode & MODE_READ) ) { if ( !(xmmregs[i].mode & MODE_READ) ) {
if (mode & MODE_READ) { if (mode & MODE_READ) {
@ -315,12 +326,10 @@ int _allocACCtoXMMreg(VURegs *VU, int xmmreg, int mode) {
return i; return i;
} }
if (xmmreg == -1) { if (xmmreg == -1)
xmmreg = _getFreeXMMreg(); xmmreg = _getFreeXMMreg();
} else
else {
_freeXMMreg(xmmreg); _freeXMMreg(xmmreg);
}
g_xmmtypes[xmmreg] = XMMT_FPS; g_xmmtypes[xmmreg] = XMMT_FPS;
xmmregs[xmmreg].inuse = 1; xmmregs[xmmreg].inuse = 1;
@ -331,9 +340,12 @@ int _allocACCtoXMMreg(VURegs *VU, int xmmreg, int mode) {
xmmregs[xmmreg].counter = g_xmmAllocCounter++; xmmregs[xmmreg].counter = g_xmmAllocCounter++;
xmmregs[xmmreg].reg = 0; xmmregs[xmmreg].reg = 0;
if (mode & MODE_READ) { if (mode & MODE_READ)
if( readfromreg >= 0 ) SSE_MOVAPS_XMM_to_XMM(xmmreg, readfromreg); {
else SSE_MOVAPS_M128_to_XMM(xmmreg, VU_ACCx_ADDR); if( readfromreg >= 0 )
SSE_MOVAPS_XMM_to_XMM(xmmreg, readfromreg);
else
SSE_MOVAPS_M128_to_XMM(xmmreg, VU_ACCx_ADDR);
} }
return xmmreg; return xmmreg;
@ -359,9 +371,7 @@ int _allocFPtoXMMreg(int xmmreg, int fpreg, int mode) {
return i; return i;
} }
if (xmmreg == -1) { if (xmmreg == -1) xmmreg = _getFreeXMMreg();
xmmreg = _getFreeXMMreg();
}
g_xmmtypes[xmmreg] = XMMT_FPS; g_xmmtypes[xmmreg] = XMMT_FPS;
xmmregs[xmmreg].inuse = 1; xmmregs[xmmreg].inuse = 1;
@ -371,9 +381,8 @@ int _allocFPtoXMMreg(int xmmreg, int fpreg, int mode) {
xmmregs[xmmreg].needed = 1; xmmregs[xmmreg].needed = 1;
xmmregs[xmmreg].counter = g_xmmAllocCounter++; xmmregs[xmmreg].counter = g_xmmAllocCounter++;
if (mode & MODE_READ) { if (mode & MODE_READ)
SSE_MOVSS_M32_to_XMM(xmmreg, (uptr)&fpuRegs.fpr[fpreg].f); SSE_MOVSS_M32_to_XMM(xmmreg, (uptr)&fpuRegs.fpr[fpreg].f);
}
return xmmreg; return xmmreg;
} }
@ -382,7 +391,8 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
{ {
int i; int i;
for (i=0; i<XMMREGS; i++) { for (i=0; i<XMMREGS; i++)
{
if (xmmregs[i].inuse == 0) continue; if (xmmregs[i].inuse == 0) continue;
if (xmmregs[i].type != XMMTYPE_GPRREG) continue; if (xmmregs[i].type != XMMTYPE_GPRREG) continue;
if (xmmregs[i].reg != gprreg) continue; if (xmmregs[i].reg != gprreg) continue;
@ -392,11 +402,14 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
#endif #endif
g_xmmtypes[i] = XMMT_INT; g_xmmtypes[i] = XMMT_INT;
if( !(xmmregs[i].mode & MODE_READ) && (mode&MODE_READ)) { if (!(xmmregs[i].mode & MODE_READ) && (mode & MODE_READ))
if( gprreg == 0 ) { {
if (gprreg == 0 )
{
SSEX_PXOR_XMM_to_XMM(i, i); SSEX_PXOR_XMM_to_XMM(i, i);
} }
else { else
{
//assert( !(g_cpuHasConstReg & (1<<gprreg)) || (g_cpuFlushedConstReg & (1<<gprreg)) ); //assert( !(g_cpuHasConstReg & (1<<gprreg)) || (g_cpuFlushedConstReg & (1<<gprreg)) );
_flushConstReg(gprreg); _flushConstReg(gprreg);
SSEX_MOVDQA_M128_to_XMM(i, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]); SSEX_MOVDQA_M128_to_XMM(i, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]);
@ -404,7 +417,8 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
xmmregs[i].mode |= MODE_READ; xmmregs[i].mode |= MODE_READ;
} }
if( (mode & MODE_WRITE) && gprreg < 32 ) { if ((mode & MODE_WRITE) && (gprreg < 32))
{
g_cpuHasConstReg &= ~(1<<gprreg); g_cpuHasConstReg &= ~(1<<gprreg);
//assert( !(g_cpuHasConstReg & (1<<gprreg)) ); //assert( !(g_cpuHasConstReg & (1<<gprreg)) );
} }
@ -416,14 +430,14 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
} }
// currently only gpr regs are const // currently only gpr regs are const
if( (mode & MODE_WRITE) && gprreg < 32 ) { // fixme - do we really need to execute this both here and in the loop?
if ((mode & MODE_WRITE) && gprreg < 32)
{
//assert( !(g_cpuHasConstReg & (1<<gprreg)) ); //assert( !(g_cpuHasConstReg & (1<<gprreg)) );
g_cpuHasConstReg &= ~(1<<gprreg); g_cpuHasConstReg &= ~(1<<gprreg);
} }
if (xmmreg == -1) { if (xmmreg == -1) xmmreg = _getFreeXMMreg();
xmmreg = _getFreeXMMreg();
}
g_xmmtypes[xmmreg] = XMMT_INT; g_xmmtypes[xmmreg] = XMMT_INT;
xmmregs[xmmreg].inuse = 1; xmmregs[xmmreg].inuse = 1;
@ -433,28 +447,35 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
xmmregs[xmmreg].needed = 1; xmmregs[xmmreg].needed = 1;
xmmregs[xmmreg].counter = g_xmmAllocCounter++; xmmregs[xmmreg].counter = g_xmmAllocCounter++;
if (mode & MODE_READ) { if (mode & MODE_READ)
if( gprreg == 0 ) { {
if (gprreg == 0 )
{
SSEX_PXOR_XMM_to_XMM(xmmreg, xmmreg); SSEX_PXOR_XMM_to_XMM(xmmreg, xmmreg);
} }
else { else
{
// DOX86 // DOX86
int mmxreg; int mmxreg;
if( (mode&MODE_READ) ) _flushConstReg(gprreg);
if (mode & MODE_READ) _flushConstReg(gprreg);
#ifndef __x86_64__ #ifndef __x86_64__
if( (mmxreg = _checkMMXreg(MMX_GPR+gprreg, 0)) >= 0 ) { mmxreg = _checkMMXreg(MMX_GPR+gprreg, 0);
// transfer
if (mmxreg >= 0 )
{
// transfer
SetMMXstate(); SetMMXstate();
SSE2_MOVQ2DQ_MM_to_XMM(xmmreg, mmxreg); SSE2_MOVQ2DQ_MM_to_XMM(xmmreg, mmxreg);
SSE2_PUNPCKLQDQ_XMM_to_XMM(xmmreg, xmmreg); SSE2_PUNPCKLQDQ_XMM_to_XMM(xmmreg, xmmreg);
SSE2_PUNPCKHQDQ_M128_to_XMM(xmmreg, (u32)&cpuRegs.GPR.r[gprreg].UL[0]); SSE2_PUNPCKHQDQ_M128_to_XMM(xmmreg, (u32)&cpuRegs.GPR.r[gprreg].UL[0]);
if( mmxregs[mmxreg].mode & MODE_WRITE ) { if (mmxregs[mmxreg].mode & MODE_WRITE )
{
// instead of setting to write, just flush to mem // instead of setting to write, just flush to mem
if( !(mode & MODE_WRITE) ) { if (!(mode & MODE_WRITE))
{
SetMMXstate(); SetMMXstate();
MOVQRtoM((u32)&cpuRegs.GPR.r[gprreg].UL[0], mmxreg); MOVQRtoM((u32)&cpuRegs.GPR.r[gprreg].UL[0], mmxreg);
} }
@ -465,27 +486,30 @@ int _allocGPRtoXMMreg(int xmmreg, int gprreg, int mode)
mmxregs[mmxreg].inuse = 0; mmxregs[mmxreg].inuse = 0;
} }
#else #else
if( (mmxreg = _checkX86reg(X86TYPE_GPR, gprreg, 0)) >= 0 ) { mmxreg = _checkX86reg(X86TYPE_GPR, gprreg, 0);
SSE2_MOVQ_R_to_XMM(xmmreg, mmxreg);
SSE_MOVHPS_M64_to_XMM(xmmreg, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]); if (mmxreg >= 0 )
{
SSE2_MOVQ_R_to_XMM(xmmreg, mmxreg);
SSE_MOVHPS_M64_to_XMM(xmmreg, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]);
// read only, instead of setting to write, just flush to mem // read only, instead of setting to write, just flush to mem
if( !(mode&MODE_WRITE) && (x86regs[mmxreg].mode & MODE_WRITE) ) { if (!(mode & MODE_WRITE) && (x86regs[mmxreg].mode & MODE_WRITE) )
MOV64RtoM((uptr)&cpuRegs.GPR.r[gprreg].UL[0], mmxreg); MOV64RtoM((uptr)&cpuRegs.GPR.r[gprreg].UL[0], mmxreg);
}
x86regs[mmxreg].inuse = 0; x86regs[mmxreg].inuse = 0;
} }
#endif #endif
else else
{ {
SSEX_MOVDQA_M128_to_XMM(xmmreg, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]); SSEX_MOVDQA_M128_to_XMM(xmmreg, (uptr)&cpuRegs.GPR.r[gprreg].UL[0]);
} }
} }
} }
else { else
{
#ifndef __x86_64__ #ifndef __x86_64__
_deleteMMXreg(MMX_GPR+gprreg, 0); _deleteMMXreg(MMX_GPR+gprreg, 0);
#else #else
_deleteX86reg(X86TYPE_GPR, gprreg, 0); _deleteX86reg(X86TYPE_GPR, gprreg, 0);
#endif #endif
@ -625,34 +649,42 @@ void _deleteVFtoXMMreg(int reg, int vu, int flush)
int i; int i;
VURegs *VU = vu ? &VU1 : &VU0; VURegs *VU = vu ? &VU1 : &VU0;
for (i=0; i<XMMREGS; i++) { for (i=0; i<XMMREGS; i++)
{
if (xmmregs[i].inuse && xmmregs[i].type == XMMTYPE_VFREG && xmmregs[i].reg == reg && xmmregs[i].VU == vu) { if (xmmregs[i].inuse && (xmmregs[i].type == XMMTYPE_VFREG) &&
(xmmregs[i].reg == reg) && (xmmregs[i].VU == vu))
{
switch(flush) { switch(flush) {
case 0: case 0:
_freeXMMreg(i); _freeXMMreg(i);
break; break;
case 1: case 1:
case 2: case 2:
if( xmmregs[i].mode & MODE_WRITE ) { if( xmmregs[i].mode & MODE_WRITE )
{
assert( reg != 0 ); assert( reg != 0 );
if( xmmregs[i].mode & MODE_VUXYZ ) { if( xmmregs[i].mode & MODE_VUXYZ )
{
if( xmmregs[i].mode & MODE_VUZ ) { if( xmmregs[i].mode & MODE_VUZ )
{
// xyz, don't destroy w // xyz, don't destroy w
int t0reg; int t0reg;
for(t0reg = 0; t0reg < XMMREGS; ++t0reg ) {
if( !xmmregs[t0reg].inuse ) break; for (t0reg = 0; t0reg < XMMREGS; ++t0reg)
{
if (!xmmregs[t0reg].inuse )
break;
} }
if( t0reg < XMMREGS ) { if (t0reg < XMMREGS )
{
SSE_MOVHLPS_XMM_to_XMM(t0reg, i); SSE_MOVHLPS_XMM_to_XMM(t0reg, i);
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i); SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i);
SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[i].reg)+8, t0reg); SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[i].reg)+8, t0reg);
} }
else { else
{
// no free reg // no free reg
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i); SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i);
SSE_SHUFPS_XMM_to_XMM(i, i, 0xc6); SSE_SHUFPS_XMM_to_XMM(i, i, 0xc6);
@ -660,7 +692,8 @@ void _deleteVFtoXMMreg(int reg, int vu, int flush)
SSE_SHUFPS_XMM_to_XMM(i, i, 0xc6); SSE_SHUFPS_XMM_to_XMM(i, i, 0xc6);
} }
} }
else { else
{
// xy // xy
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i); SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[i].reg), i);
} }
@ -672,8 +705,7 @@ void _deleteVFtoXMMreg(int reg, int vu, int flush)
xmmregs[i].mode |= MODE_READ; xmmregs[i].mode |= MODE_READ;
} }
if( flush == 2 ) if (flush == 2) xmmregs[i].inuse = 0;
xmmregs[i].inuse = 0;
break; break;
} }
@ -688,8 +720,7 @@ void _deleteACCtoXMMreg(int vu, int flush)
VURegs *VU = vu ? &VU1 : &VU0; VURegs *VU = vu ? &VU1 : &VU0;
for (i=0; i<XMMREGS; i++) { for (i=0; i<XMMREGS; i++) {
if (xmmregs[i].inuse && (xmmregs[i].type == XMMTYPE_ACC) && (xmmregs[i].VU == vu)) {
if (xmmregs[i].inuse && xmmregs[i].type == XMMTYPE_ACC && xmmregs[i].VU == vu) {
switch(flush) { switch(flush) {
case 0: case 0:
@ -806,84 +837,110 @@ void _deleteFPtoXMMreg(int reg, int flush)
} }
} }
void _freeXMMreg(int xmmreg) { void _freeXMMreg(int xmmreg)
{
VURegs *VU = xmmregs[xmmreg].VU ? &VU1 : &VU0; VURegs *VU = xmmregs[xmmreg].VU ? &VU1 : &VU0;
assert( xmmreg < XMMREGS ); assert( xmmreg < XMMREGS );
if (!xmmregs[xmmreg].inuse) return; if (!xmmregs[xmmreg].inuse) return;
if (xmmregs[xmmreg].type == XMMTYPE_VFREG && (xmmregs[xmmreg].mode & MODE_WRITE) ) { if (xmmregs[xmmreg].mode & MODE_WRITE) {
if( xmmregs[xmmreg].mode & MODE_VUXYZ ) { switch (xmmregs[xmmreg].type) {
case XMMTYPE_VFREG:
if( xmmregs[xmmreg].mode & MODE_VUXYZ )
{
if( xmmregs[xmmreg].mode & MODE_VUZ )
{
// don't destroy w
int t0reg;
for(t0reg = 0; t0reg < XMMREGS; ++t0reg ) {
if( !xmmregs[t0reg].inuse ) break;
}
if( xmmregs[xmmreg].mode & MODE_VUZ ) { if( t0reg < XMMREGS )
// don't destroy w {
int t0reg; SSE_MOVHLPS_XMM_to_XMM(t0reg, xmmreg);
for(t0reg = 0; t0reg < XMMREGS; ++t0reg ) { SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
if( !xmmregs[t0reg].inuse ) break; SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[xmmreg].reg)+8, t0reg);
}
else
{
// no free reg
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
//SSE_MOVHLPS_XMM_to_XMM(xmmreg, xmmreg);
SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[xmmreg].reg)+8, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
}
} }
else
if( t0reg < XMMREGS ) { {
SSE_MOVHLPS_XMM_to_XMM(t0reg, xmmreg);
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg); SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[xmmreg].reg)+8, t0reg);
}
else {
// no free reg
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
//SSE_MOVHLPS_XMM_to_XMM(xmmreg, xmmreg);
SSE_MOVSS_XMM_to_M32(VU_VFx_ADDR(xmmregs[xmmreg].reg)+8, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
} }
} }
else { else
SSE_MOVLPS_XMM_to_M64(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg); {
SSE_MOVAPS_XMM_to_M128(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
} }
} break;
else SSE_MOVAPS_XMM_to_M128(VU_VFx_ADDR(xmmregs[xmmreg].reg), xmmreg);
} case XMMTYPE_ACC:
else if (xmmregs[xmmreg].type == XMMTYPE_ACC && (xmmregs[xmmreg].mode & MODE_WRITE) ) { if( xmmregs[xmmreg].mode & MODE_VUXYZ )
if( xmmregs[xmmreg].mode & MODE_VUXYZ ) { {
if( xmmregs[xmmreg].mode & MODE_VUZ )
{
// don't destroy w
int t0reg;
for(t0reg = 0; t0reg < XMMREGS; ++t0reg ) {
if( !xmmregs[t0reg].inuse ) break;
}
if( xmmregs[xmmreg].mode & MODE_VUZ ) { if( t0reg < XMMREGS )
// don't destroy w {
int t0reg; SSE_MOVHLPS_XMM_to_XMM(t0reg, xmmreg);
for(t0reg = 0; t0reg < XMMREGS; ++t0reg ) { SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg);
if( !xmmregs[t0reg].inuse ) break; SSE_MOVSS_XMM_to_M32(VU_ACCx_ADDR+8, t0reg);
}
else
{
// no free reg
SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
//SSE_MOVHLPS_XMM_to_XMM(xmmreg, xmmreg);
SSE_MOVSS_XMM_to_M32(VU_ACCx_ADDR+8, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
}
} }
else
if( t0reg < XMMREGS ) { {
SSE_MOVHLPS_XMM_to_XMM(t0reg, xmmreg);
SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg); SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg);
SSE_MOVSS_XMM_to_M32(VU_ACCx_ADDR+8, t0reg);
}
else {
// no free reg
SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
//SSE_MOVHLPS_XMM_to_XMM(xmmreg, xmmreg);
SSE_MOVSS_XMM_to_M32(VU_ACCx_ADDR+8, xmmreg);
SSE_SHUFPS_XMM_to_XMM(xmmreg, xmmreg, 0xc6);
} }
} }
else { else
SSE_MOVLPS_XMM_to_M64(VU_ACCx_ADDR, xmmreg); {
SSE_MOVAPS_XMM_to_M128(VU_ACCx_ADDR, xmmreg);
} }
} break;
else SSE_MOVAPS_XMM_to_M128(VU_ACCx_ADDR, xmmreg);
case XMMTYPE_GPRREG:
assert( xmmregs[xmmreg].reg != 0 );
//assert( g_xmmtypes[xmmreg] == XMMT_INT );
SSEX_MOVDQA_XMM_to_M128((uptr)&cpuRegs.GPR.r[xmmregs[xmmreg].reg].UL[0], xmmreg);
break;
case XMMTYPE_FPREG:
SSE_MOVSS_XMM_to_M32((uptr)&fpuRegs.fpr[xmmregs[xmmreg].reg], xmmreg);
break;
case XMMTYPE_FPACC:
SSE_MOVSS_XMM_to_M32((uptr)&fpuRegs.ACC.f, xmmreg);
break;
default:
break;
} }
else if (xmmregs[xmmreg].type == XMMTYPE_GPRREG && (xmmregs[xmmreg].mode & MODE_WRITE) ) {
assert( xmmregs[xmmreg].reg != 0 );
//assert( g_xmmtypes[xmmreg] == XMMT_INT );
SSEX_MOVDQA_XMM_to_M128((uptr)&cpuRegs.GPR.r[xmmregs[xmmreg].reg].UL[0], xmmreg);
} }
else if (xmmregs[xmmreg].type == XMMTYPE_FPREG && (xmmregs[xmmreg].mode & MODE_WRITE)) {
SSE_MOVSS_XMM_to_M32((uptr)&fpuRegs.fpr[xmmregs[xmmreg].reg], xmmreg);
}
else if (xmmregs[xmmreg].type == XMMTYPE_FPACC && (xmmregs[xmmreg].mode & MODE_WRITE)) {
SSE_MOVSS_XMM_to_M32((uptr)&fpuRegs.ACC.f, xmmreg);
}
xmmregs[xmmreg].mode &= ~(MODE_WRITE|MODE_VUXYZ); xmmregs[xmmreg].mode &= ~(MODE_WRITE|MODE_VUXYZ);
xmmregs[xmmreg].inuse = 0; xmmregs[xmmreg].inuse = 0;
} }
@ -1184,12 +1241,13 @@ void _recClearInst(EEINST* pinst)
} }
// returns nonzero value if reg has been written between [startpc, endpc-4] // returns nonzero value if reg has been written between [startpc, endpc-4]
int _recIsRegWritten(EEINST* pinst, int size, u8 xmmtype, u8 reg) u32 _recIsRegWritten(EEINST* pinst, int size, u8 xmmtype, u8 reg)
{ {
int i, inst = 1; u32 i, inst = 1;
while(size-- > 0) { while(size-- > 0) {
for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) { for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) {
if( pinst->writeType[i] == xmmtype && pinst->writeReg[i] == reg ) if ((pinst->writeType[i] == xmmtype) && (pinst->writeReg[i] == reg))
return inst; return inst;
} }
++inst; ++inst;
@ -1199,9 +1257,9 @@ int _recIsRegWritten(EEINST* pinst, int size, u8 xmmtype, u8 reg)
return 0; return 0;
} }
int _recIsRegUsed(EEINST* pinst, int size, u8 xmmtype, u8 reg) u32 _recIsRegUsed(EEINST* pinst, int size, u8 xmmtype, u8 reg)
{ {
int i, inst = 1; u32 i, inst = 1;
while(size-- > 0) { while(size-- > 0) {
for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) { for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) {
if( pinst->writeType[i] == xmmtype && pinst->writeReg[i] == reg ) if( pinst->writeType[i] == xmmtype && pinst->writeReg[i] == reg )
@ -1220,8 +1278,8 @@ int _recIsRegUsed(EEINST* pinst, int size, u8 xmmtype, u8 reg)
void _recFillRegister(EEINST* pinst, int type, int reg, int write) void _recFillRegister(EEINST* pinst, int type, int reg, int write)
{ {
int i = 0; u32 i = 0;
if( write ) { if (write ) {
for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) { for(i = 0; i < ARRAYSIZE(pinst->writeType); ++i) {
if( pinst->writeType[i] == XMMTYPE_TEMP ) { if( pinst->writeType[i] == XMMTYPE_TEMP ) {
pinst->writeType[i] = type; pinst->writeType[i] = type;

View File

@ -323,9 +323,9 @@ extern EEINST* g_pCurInstInfo; // info for the cur instruction
void _recClearInst(EEINST* pinst); void _recClearInst(EEINST* pinst);
// returns the number of insts + 1 until written (0 if not written) // returns the number of insts + 1 until written (0 if not written)
int _recIsRegWritten(EEINST* pinst, int size, u8 xmmtype, u8 reg); u32 _recIsRegWritten(EEINST* pinst, int size, u8 xmmtype, u8 reg);
// returns the number of insts + 1 until used (0 if not used) // returns the number of insts + 1 until used (0 if not used)
int _recIsRegUsed(EEINST* pinst, int size, u8 xmmtype, u8 reg); u32 _recIsRegUsed(EEINST* pinst, int size, u8 xmmtype, u8 reg);
void _recFillRegister(EEINST* pinst, int type, int reg, int write); void _recFillRegister(EEINST* pinst, int type, int reg, int write);
#define EEINST_ISLIVE64(reg) (g_pCurInstInfo->regs[reg] & (EEINST_LIVE0|EEINST_LIVE1)) #define EEINST_ISLIVE64(reg) (g_pCurInstInfo->regs[reg] & (EEINST_LIVE0|EEINST_LIVE1))