Glide64: Update to make asm more closer to original code
This commit is contained in:
zilmar
parent
047eab9b70
commit
8f16f58311
|
|
@ -203,400 +203,17 @@ TRANSFORMVECTOR InverseTransformVector = InverseTransformVectorC;
|
||||||
DOTPRODUCT DotProduct = DotProductC;
|
DOTPRODUCT DotProduct = DotProductC;
|
||||||
NORMALIZEVECTOR NormalizeVector = NormalizeVectorC;
|
NORMALIZEVECTOR NormalizeVector = NormalizeVectorC;
|
||||||
|
|
||||||
void TransformVectorSSE(float *src, float *dst, float mat[4][4])
|
extern "C" void TransformVectorSSE(float *src, float *dst, float mat[4][4]);
|
||||||
{
|
extern "C" void TransformVector3DNOW(float *src, float *dst, float mat[4][4]);
|
||||||
__asm
|
extern "C" void InverseTransformVector3DNOW(float *src, float *dst, float mat[4][4]);
|
||||||
{
|
extern "C" void MulMatricesSSE(float m1[4][4],float m2[4][4],float r[4][4]);
|
||||||
mov ecx,[src]
|
extern "C" void MulMatrices3DNOW(float m1[4][4],float m2[4][4],float r[4][4]);
|
||||||
mov eax,[dst]
|
extern "C" float DotProductSSE3(register float *v1, register float *v2);
|
||||||
mov edx,[mat]
|
|
||||||
|
|
||||||
movss xmm0,[ecx] ; 0 0 0 src[0]
|
|
||||||
movss xmm5,[edx] ; 0 0 0 mat[0][0]
|
|
||||||
movhps xmm5,[edx+4] ; mat[0][2] mat[0][1] 0 mat[0][0]
|
|
||||||
shufps xmm0,xmm0, 0 ; src[0] src[0] src[0] src[0]
|
|
||||||
movss xmm1,[ecx+4] ; 0 0 0 src[1]
|
|
||||||
movss xmm3,[edx+16] ; 0 0 0 mat[1][0]
|
|
||||||
movhps xmm3,[edx+20] ; mat[1][2] mat[1][1] 0 mat[1][0]
|
|
||||||
shufps xmm1,xmm1, 0 ; src[1] src[1] src[1] src[1]
|
|
||||||
mulps xmm0,xmm5 ; mat[0][2]*src[0] mat[0][1]*src[0] 0 mat[0][0]*src[0]
|
|
||||||
mulps xmm1,xmm3 ; mat[1][2]*src[1] mat[1][1]*src[1] 0 mat[1][0]*src[1]
|
|
||||||
movss xmm2,[ecx+8] ; 0 0 0 src[2]
|
|
||||||
shufps xmm2,xmm2, 0 ; src[2] src[2] src[2] src[2]
|
|
||||||
movss xmm4,[edx+32] ; 0 0 0 mat[2][0]
|
|
||||||
movhps xmm4,[edx+36] ; mat[2][2] mat[2][1] 0 mat[2][0]
|
|
||||||
addps xmm0,xmm1 ; mat[0][2]*src[0]+mat[1][2]*src[1] mat[0][1]*src[0]+mat[1][1]*src[1] 0 mat[0][0]*src[0]+mat[1][0]*src[1]
|
|
||||||
mulps xmm2,xmm4 ; mat[2][2]*src[2] mat[2][1]*src[2] 0 mat[2][0]*src[2]
|
|
||||||
addps xmm0,xmm2 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2] mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2] 0 mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
|
||||||
movss [eax],xmm0 ; mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
|
||||||
movhps [eax+4],xmm0 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2] mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2]
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void TransformVector3DNOW(float *src, float *dst, float mat[4][4])
|
|
||||||
{
|
|
||||||
_asm {
|
|
||||||
femms
|
|
||||||
mov ecx,[src]
|
|
||||||
mov eax,[dst]
|
|
||||||
mov edx,[mat]
|
|
||||||
movq mm0,[ecx] ; src[1] src[0]
|
|
||||||
movd mm2,[ecx+8] ; 0 src[2]
|
|
||||||
movq mm1,mm0 ; src[1] src[0]
|
|
||||||
punpckldq mm0,mm0 ; src[0] src[0]
|
|
||||||
punpckhdq mm1,mm1 ; src[1] src[1]
|
|
||||||
punpckldq mm2,mm2 ; src[2] src[2]
|
|
||||||
movq mm3,mm0 ; src[0] src[0]
|
|
||||||
movq mm4,mm1 ; src[1] src[1]
|
|
||||||
movq mm5,mm2 ; src[2] src[2]
|
|
||||||
pfmul mm0,[edx] ; src[0]*mat[0][1] src[0]*mat[0][0]
|
|
||||||
pfmul mm3,[edx+8] ; 0 src[0]*mat[0][2]
|
|
||||||
pfmul mm1,[edx+16] ; src[1]*mat[1][1] src[1]*mat[1][0]
|
|
||||||
pfmul mm4,[edx+24] ; 0 src[1]*mat[1][2]
|
|
||||||
pfmul mm2,[edx+32] ; src[2]*mat[2][1] src[2]*mat[2][0]
|
|
||||||
pfmul mm5,[edx+40] ; 0 src[2]*mat[2][2]
|
|
||||||
pfadd mm0,mm1 ; src[0]*mat[0][1]+src[1]*mat[1][1] src[0]*mat[0][0]+src[1]*mat[1][0]
|
|
||||||
pfadd mm3,mm4 ; 0 src[0]*mat[0][2]+src[1]*mat[1][2]
|
|
||||||
pfadd mm0,mm2 ; src[0]*mat[0][1]+src[1]*mat[1][1]+src[2]*mat[2][1] src[0]*mat[0][0]+src[1]*mat[1][0]+src[2]*mat[2][0]
|
|
||||||
pfadd mm3,mm5 ; 0 src[0]*mat[0][2]+src[1]*mat[1][2]+src[2]*mat[2][2]
|
|
||||||
movq [eax],mm0 ; mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2] mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
|
||||||
movd [eax+8],mm3 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2]
|
|
||||||
femms
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void InverseTransformVector3DNOW(float *src, float *dst, float mat[4][4])
|
|
||||||
{
|
|
||||||
_asm {
|
|
||||||
femms
|
|
||||||
mov ecx,[src]
|
|
||||||
mov eax,[dst]
|
|
||||||
mov edx,[mat]
|
|
||||||
movq mm0,[ecx] ; src[1] src[0]
|
|
||||||
movd mm4,[ecx+8] ; 0 src[2]
|
|
||||||
movq mm1,mm0 ; src[1] src[0]
|
|
||||||
pfmul mm0,[edx] ; src[1]*mat[0][1] src[0]*mat[0][0]
|
|
||||||
movq mm5,mm4 ; 0 src[2]
|
|
||||||
pfmul mm4,[edx+8] ; 0 src[2]*mat[0][2]
|
|
||||||
movq mm2,mm1 ; src[1] src[0]
|
|
||||||
pfmul mm1,[edx+16] ; src[1]*mat[1][1] src[0]*mat[1][0]
|
|
||||||
movq mm6,mm5 ; 0 src[2]
|
|
||||||
pfmul mm5,[edx+24] ; 0 src[2]*mat[1][2]
|
|
||||||
movq mm3,mm2 ; src[1] src[0]
|
|
||||||
pfmul mm2,[edx+32] ; src[1]*mat[2][1] src[0]*mat[2][0]
|
|
||||||
movq mm7,mm6 ; 0 src[2]
|
|
||||||
pfmul mm6,[edx+40] ; 0 src[2]*mat[2][2]
|
|
||||||
pfacc mm0,mm4 ; src[2]*mat[0][2] src[1]*mat[0][1]+src[0]*mat[0][0]
|
|
||||||
pfacc mm1,mm5 ; src[2]*mat[1][2] src[1]*mat[1][1]+src[0]*mat[1][0]
|
|
||||||
pfacc mm2,mm6 ; src[2]*mat[2][2] src[1]*mat[2][1]+src[0]*mat[2][0]
|
|
||||||
pfacc mm0,mm1 ; src[2]*mat[1][2]+src[1]*mat[1][1]+src[0]*mat[1][0] src[2]*mat[0][2]+src[1]*mat[0][1]+src[0]*mat[0][0]
|
|
||||||
pfacc mm2,mm3 ; 0 src[2]*mat[2][2]+src[1]*mat[2][1]+src[0]*mat[2][0]
|
|
||||||
movq [eax],mm0 ; mat[1][0]*src[0]+mat[1][1]*src[1]+mat[1][2]*src[2] mat[0][0]*src[0]+mat[0][1]*src[1]+mat[0][2]*src[2]
|
|
||||||
movd [eax+8],mm2 ; mat[2][0]*src[0]+mat[2][1]*src[1]+mat[2][2]*src[2]
|
|
||||||
femms
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void MulMatricesSSE(float m1[4][4],float m2[4][4],float r[4][4])
|
|
||||||
{
|
|
||||||
__asm
|
|
||||||
{
|
|
||||||
mov eax,[r]
|
|
||||||
mov ecx,[m1]
|
|
||||||
mov edx,[m2]
|
|
||||||
|
|
||||||
movaps xmm0,[edx]
|
|
||||||
movaps xmm1,[edx+16]
|
|
||||||
movaps xmm2,[edx+32]
|
|
||||||
movaps xmm3,[edx+48]
|
|
||||||
|
|
||||||
; r[0][0],r[0][1],r[0][2],r[0][3]
|
|
||||||
|
|
||||||
movaps xmm4,[ecx]
|
|
||||||
movaps xmm5,xmm4
|
|
||||||
movaps xmm6,xmm4
|
|
||||||
movaps xmm7,xmm4
|
|
||||||
|
|
||||||
shufps xmm4,xmm4,00000000b
|
|
||||||
shufps xmm5,xmm5,01010101b
|
|
||||||
shufps xmm6,xmm6,10101010b
|
|
||||||
shufps xmm7,xmm7,11111111b
|
|
||||||
|
|
||||||
mulps xmm4,xmm0
|
|
||||||
mulps xmm5,xmm1
|
|
||||||
mulps xmm6,xmm2
|
|
||||||
mulps xmm7,xmm3
|
|
||||||
|
|
||||||
addps xmm4,xmm5
|
|
||||||
addps xmm4,xmm6
|
|
||||||
addps xmm4,xmm7
|
|
||||||
|
|
||||||
movaps [eax],xmm4
|
|
||||||
|
|
||||||
; r[1][0],r[1][1],r[1][2],r[1][3]
|
|
||||||
|
|
||||||
movaps xmm4,[ecx+16]
|
|
||||||
movaps xmm5,xmm4
|
|
||||||
movaps xmm6,xmm4
|
|
||||||
movaps xmm7,xmm4
|
|
||||||
|
|
||||||
shufps xmm4,xmm4,00000000b
|
|
||||||
shufps xmm5,xmm5,01010101b
|
|
||||||
shufps xmm6,xmm6,10101010b
|
|
||||||
shufps xmm7,xmm7,11111111b
|
|
||||||
|
|
||||||
mulps xmm4,xmm0
|
|
||||||
mulps xmm5,xmm1
|
|
||||||
mulps xmm6,xmm2
|
|
||||||
mulps xmm7,xmm3
|
|
||||||
|
|
||||||
addps xmm4,xmm5
|
|
||||||
addps xmm4,xmm6
|
|
||||||
addps xmm4,xmm7
|
|
||||||
|
|
||||||
movaps [eax+16],xmm4
|
|
||||||
|
|
||||||
|
|
||||||
; r[2][0],r[2][1],r[2][2],r[2][3]
|
|
||||||
|
|
||||||
movaps xmm4,[ecx+32]
|
|
||||||
movaps xmm5,xmm4
|
|
||||||
movaps xmm6,xmm4
|
|
||||||
movaps xmm7,xmm4
|
|
||||||
|
|
||||||
shufps xmm4,xmm4,00000000b
|
|
||||||
shufps xmm5,xmm5,01010101b
|
|
||||||
shufps xmm6,xmm6,10101010b
|
|
||||||
shufps xmm7,xmm7,11111111b
|
|
||||||
|
|
||||||
mulps xmm4,xmm0
|
|
||||||
mulps xmm5,xmm1
|
|
||||||
mulps xmm6,xmm2
|
|
||||||
mulps xmm7,xmm3
|
|
||||||
|
|
||||||
addps xmm4,xmm5
|
|
||||||
addps xmm4,xmm6
|
|
||||||
addps xmm4,xmm7
|
|
||||||
|
|
||||||
movaps [eax+32],xmm4
|
|
||||||
|
|
||||||
; r[3][0],r[3][1],r[3][2],r[3][3]
|
|
||||||
|
|
||||||
movaps xmm4,[ecx+48]
|
|
||||||
movaps xmm5,xmm4
|
|
||||||
movaps xmm6,xmm4
|
|
||||||
movaps xmm7,xmm4
|
|
||||||
|
|
||||||
shufps xmm4,xmm4,00000000b
|
|
||||||
shufps xmm5,xmm5,01010101b
|
|
||||||
shufps xmm6,xmm6,10101010b
|
|
||||||
shufps xmm7,xmm7,11111111b
|
|
||||||
|
|
||||||
mulps xmm4,xmm0
|
|
||||||
mulps xmm5,xmm1
|
|
||||||
mulps xmm6,xmm2
|
|
||||||
mulps xmm7,xmm3
|
|
||||||
|
|
||||||
addps xmm4,xmm5
|
|
||||||
addps xmm4,xmm6
|
|
||||||
addps xmm4,xmm7
|
|
||||||
|
|
||||||
movaps [eax+48],xmm4
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void MulMatrices3DNOW(float m1[4][4],float m2[4][4],float r[4][4])
|
|
||||||
{
|
|
||||||
_asm
|
|
||||||
{
|
|
||||||
femms
|
|
||||||
mov ecx,[m1]
|
|
||||||
mov eax,[r]
|
|
||||||
mov edx,[m2]
|
|
||||||
|
|
||||||
movq mm0,[ecx]
|
|
||||||
movq mm1,[ecx+8]
|
|
||||||
movq mm4,[edx]
|
|
||||||
punpckhdq mm2,mm0
|
|
||||||
movq mm5,[edx+16]
|
|
||||||
punpckhdq mm3,mm1
|
|
||||||
movq mm6,[edx+32]
|
|
||||||
punpckldq mm0,mm0
|
|
||||||
punpckldq mm1,mm1
|
|
||||||
pfmul mm4,mm0
|
|
||||||
punpckhdq mm2,mm2
|
|
||||||
pfmul mm0,[edx+8]
|
|
||||||
movq mm7,[edx+48]
|
|
||||||
pfmul mm5,mm2
|
|
||||||
punpckhdq mm3,mm3
|
|
||||||
pfmul mm2,[edx+24]
|
|
||||||
pfmul mm6,mm1
|
|
||||||
pfadd mm5,mm4
|
|
||||||
pfmul mm1,[edx+40]
|
|
||||||
pfadd mm2,mm0
|
|
||||||
pfmul mm7,mm3
|
|
||||||
pfadd mm6,mm5
|
|
||||||
pfmul mm3,[edx+56]
|
|
||||||
pfadd mm2,mm1
|
|
||||||
pfadd mm7,mm6
|
|
||||||
movq mm0,[ecx+16]
|
|
||||||
pfadd mm3,mm2
|
|
||||||
movq mm1,[ecx+24]
|
|
||||||
movq [eax],mm7
|
|
||||||
movq mm4,[edx]
|
|
||||||
movq [eax+8],mm3
|
|
||||||
|
|
||||||
punpckhdq mm2,mm0
|
|
||||||
movq mm5,[edx+16]
|
|
||||||
punpckhdq mm3,mm1
|
|
||||||
movq mm6,[edx+32]
|
|
||||||
punpckldq mm0,mm0
|
|
||||||
punpckldq mm1,mm1
|
|
||||||
pfmul mm4,mm0
|
|
||||||
punpckhdq mm2,mm2
|
|
||||||
pfmul mm0,[edx+8]
|
|
||||||
movq mm7,[edx+48]
|
|
||||||
pfmul mm5,mm2
|
|
||||||
punpckhdq mm3,mm3
|
|
||||||
pfmul mm2,[edx+24]
|
|
||||||
pfmul mm6,mm1
|
|
||||||
pfadd mm5,mm4
|
|
||||||
pfmul mm1,[edx+40]
|
|
||||||
pfadd mm2,mm0
|
|
||||||
pfmul mm7,mm3
|
|
||||||
pfadd mm6,mm5
|
|
||||||
pfmul mm3,[edx+56]
|
|
||||||
pfadd mm2,mm1
|
|
||||||
pfadd mm7,mm6
|
|
||||||
movq mm0,[ecx+32]
|
|
||||||
pfadd mm3,mm2
|
|
||||||
movq mm1,[ecx+40]
|
|
||||||
movq [eax+16],mm7
|
|
||||||
movq mm4,[edx]
|
|
||||||
movq [eax+24],mm3
|
|
||||||
|
|
||||||
punpckhdq mm2,mm0
|
|
||||||
movq mm5,[edx+16]
|
|
||||||
punpckhdq mm3,mm1
|
|
||||||
movq mm6,[edx+32]
|
|
||||||
punpckldq mm0,mm0
|
|
||||||
punpckldq mm1,mm1
|
|
||||||
pfmul mm4,mm0
|
|
||||||
punpckhdq mm2,mm2
|
|
||||||
pfmul mm0,[edx+8]
|
|
||||||
movq mm7,[edx+48]
|
|
||||||
pfmul mm5,mm2
|
|
||||||
punpckhdq mm3,mm3
|
|
||||||
pfmul mm2,[edx+24]
|
|
||||||
pfmul mm6,mm1
|
|
||||||
pfadd mm5,mm4
|
|
||||||
pfmul mm1,[edx+40]
|
|
||||||
pfadd mm2,mm0
|
|
||||||
pfmul mm7,mm3
|
|
||||||
pfadd mm6,mm5
|
|
||||||
pfmul mm3,[edx+56]
|
|
||||||
pfadd mm2,mm1
|
|
||||||
pfadd mm7,mm6
|
|
||||||
movq mm0,[ecx+48]
|
|
||||||
pfadd mm3,mm2
|
|
||||||
movq mm1,[ecx+56]
|
|
||||||
movq [eax+32],mm7
|
|
||||||
movq mm4,[edx]
|
|
||||||
movq [eax+40],mm3
|
|
||||||
|
|
||||||
punpckhdq mm2,mm0
|
|
||||||
movq mm5,[edx+16]
|
|
||||||
punpckhdq mm3,mm1
|
|
||||||
movq mm6,[edx+32]
|
|
||||||
punpckldq mm0,mm0
|
|
||||||
punpckldq mm1,mm1
|
|
||||||
pfmul mm4,mm0
|
|
||||||
punpckhdq mm2,mm2
|
|
||||||
pfmul mm0,[edx+8]
|
|
||||||
movq mm7,[edx+48]
|
|
||||||
pfmul mm5,mm2
|
|
||||||
punpckhdq mm3,mm3
|
|
||||||
pfmul mm2,[edx+24]
|
|
||||||
pfmul mm6,mm1
|
|
||||||
pfadd mm5,mm4
|
|
||||||
pfmul mm1,[edx+40]
|
|
||||||
pfadd mm2,mm0
|
|
||||||
pfmul mm7,mm3
|
|
||||||
pfadd mm6,mm5
|
|
||||||
pfmul mm3,[edx+56]
|
|
||||||
pfadd mm2,mm1
|
|
||||||
pfadd mm7,mm6
|
|
||||||
pfadd mm3,mm2
|
|
||||||
movq [eax+48],mm7
|
|
||||||
movq [eax+56],mm3
|
|
||||||
femms
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
float DotProductSSE3(register float *v1, register float *v2)
|
|
||||||
{
|
|
||||||
_asm {
|
|
||||||
mov eax,[v1]
|
|
||||||
mov edx,[v2]
|
|
||||||
movaps xmm0, [eax]
|
|
||||||
mulps xmm0, [edx]
|
|
||||||
haddps xmm0, xmm0
|
|
||||||
haddps xmm0, xmm0
|
|
||||||
; movss eax, xmm0
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
extern "C" float DotProduct3DNOW(register float *v1, register float *v2);
|
extern "C" float DotProduct3DNOW(register float *v1, register float *v2);
|
||||||
extern "C" void NormalizeVectorSSE(float *v);
|
extern "C" void NormalizeVectorSSE(float *v);
|
||||||
|
extern "C" void NormalizeVector3DNOW(float *v);
|
||||||
|
|
||||||
void NormalizeVector3DNOW(float *v)
|
extern "C" void DetectSIMD(int function, int * iedx, int * iecx);
|
||||||
{
|
|
||||||
_asm{
|
|
||||||
femms
|
|
||||||
mov edx,[v]
|
|
||||||
movq mm0,[edx]
|
|
||||||
movq mm3,[edx+8]
|
|
||||||
movq mm1,mm0
|
|
||||||
movq mm2,mm3
|
|
||||||
pfmul mm0,mm0
|
|
||||||
pfmul mm3,mm3
|
|
||||||
pfacc mm0,mm0
|
|
||||||
pfadd mm0,mm3
|
|
||||||
;movq mm4,mm0 ; prepare for 24bit precision
|
|
||||||
;punpckldq mm4,mm4 ; prepare for 24bit precision
|
|
||||||
pfrsqrt mm0,mm0 ; 15bit precision 1/sqrtf(v)
|
|
||||||
;movq mm3,mm0
|
|
||||||
;pfmul mm0,mm0
|
|
||||||
;pfrsqit1 mm0,mm4
|
|
||||||
;pfrcpit2 mm0,mm3 ; 24bit precision 1/sqrtf(v)
|
|
||||||
pfmul mm1,mm0
|
|
||||||
pfmul mm2,mm0
|
|
||||||
movq [edx],mm1
|
|
||||||
movq [edx+8],mm2
|
|
||||||
femms
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void DetectSIMD(int func, int * iedx, int * iecx)
|
|
||||||
{
|
|
||||||
unsigned long reg, reg2;
|
|
||||||
__asm
|
|
||||||
{
|
|
||||||
mov eax, func
|
|
||||||
cpuid
|
|
||||||
mov reg, edx
|
|
||||||
mov reg2, ecx
|
|
||||||
}
|
|
||||||
|
|
||||||
if (iedx)
|
|
||||||
{
|
|
||||||
*iedx = reg;
|
|
||||||
}
|
|
||||||
if (iecx)
|
|
||||||
{
|
|
||||||
*iecx = reg2;
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
void math_init()
|
void math_init()
|
||||||
{
|
{
|
||||||
|
|
|
||||||
|
|
@ -1,109 +0,0 @@
|
||||||
;/*
|
|
||||||
;* Glide64 - Glide video plugin for Nintendo 64 emulators.
|
|
||||||
;*
|
|
||||||
;* This program is free software; you can redistribute it and/or modify
|
|
||||||
;* it under the terms of the GNU General Public License as published by
|
|
||||||
;* the Free Software Foundation; either version 2 of the License, or
|
|
||||||
;* any later version.
|
|
||||||
;*
|
|
||||||
;* This program is distributed in the hope that it will be useful,
|
|
||||||
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
||||||
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
||||||
;* GNU General Public License for more details.
|
|
||||||
;*
|
|
||||||
;* You should have received a copy of the GNU General Public License
|
|
||||||
;* along with this program; if not, write to the Free Software
|
|
||||||
;* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
|
||||||
;*/
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; Glide64 - Glide Plugin for Nintendo 64 emulators
|
|
||||||
; Project started on December 29th, 2001
|
|
||||||
;
|
|
||||||
; Authors:
|
|
||||||
; Dave2001, original author, founded the project in 2001, left it in 2002
|
|
||||||
; Gugaman, joined the project in 2002, left it in 2002
|
|
||||||
; Sergey 'Gonetz' Lipski, joined the project in 2002, main author since fall of 2002
|
|
||||||
; Hiroshi 'KoolSmoky' Morii, joined the project in 2007
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; To modify Glide64:
|
|
||||||
; * Write your name and (optional)email, commented by your work, so I know who did it, and so that you can find which parts you modified when it comes time to send it to me.
|
|
||||||
; * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
|
|
||||||
%include "inc/c32.mac"
|
|
||||||
|
|
||||||
segment .text
|
|
||||||
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; ******** SSE ********
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
|
|
||||||
|
|
||||||
proc MulMatricesSSE
|
|
||||||
CPU P3
|
|
||||||
endproc ;MulMatricesSSE
|
|
||||||
|
|
||||||
proc NormalizeVectorSSE
|
|
||||||
CPU P3
|
|
||||||
%$v arg
|
|
||||||
|
|
||||||
mov edx, [ebp + %$v]
|
|
||||||
movaps xmm0, [edx] ; x y z 0
|
|
||||||
movaps xmm2, xmm0 ; x y z 0
|
|
||||||
mulps xmm0, xmm0 ; x*x y*y z*z 0
|
|
||||||
movaps xmm1, xmm0 ; x*x y*y z*z 0
|
|
||||||
shufps xmm0, xmm1, 0x4e ; z*z 0 x*x y*y
|
|
||||||
addps xmm0, xmm1 ; x*x+z*z y*y z*z+x*x y*y
|
|
||||||
movaps xmm1, xmm0 ; x*x+z*z y*y z*z+x*x y*y
|
|
||||||
shufps xmm1, xmm1, 0x11 ; y*y z*z+x*x y*y z*z+x*x
|
|
||||||
addps xmm0, xmm1 ; x*x+z*z+y*y
|
|
||||||
rsqrtps xmm0, xmm0 ; 1.0/sqrt(x*x+z*z+y*y)
|
|
||||||
mulps xmm2, xmm0 ; x/sqrt(x*x+z*z+y*y) y/sqrt(x*x+z*z+y*y) z/sqrt(x*x+z*z+y*y) 0
|
|
||||||
movaps [edx], xmm2
|
|
||||||
|
|
||||||
endproc ;NormalizeVectorSSE
|
|
||||||
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; ******** SSE3 ********
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
|
|
||||||
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; ******** 3DNOW ********
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
proc DotProduct3DNOW
|
|
||||||
CPU 586
|
|
||||||
%$v1 arg
|
|
||||||
%$v2 arg
|
|
||||||
|
|
||||||
femms
|
|
||||||
mov edx,[ebp + %$v1]
|
|
||||||
mov eax,[ebp + %$v2]
|
|
||||||
movq mm0,[edx]
|
|
||||||
movq mm3,[eax]
|
|
||||||
pfmul mm0,mm3
|
|
||||||
movq mm2,[edx+8]
|
|
||||||
movq mm1,[eax+8]
|
|
||||||
pfacc mm0,mm0
|
|
||||||
pfmul mm1,mm2
|
|
||||||
pfadd mm0,mm1
|
|
||||||
movd eax,mm0
|
|
||||||
femms
|
|
||||||
|
|
||||||
endproc ;DotProduct3DNOW
|
|
||||||
|
|
||||||
|
|
@ -0,0 +1,527 @@
|
||||||
|
;/*
|
||||||
|
;* Glide64 - Glide video plugin for Nintendo 64 emulators.
|
||||||
|
;*
|
||||||
|
;* This program is free software; you can redistribute it and/or modify
|
||||||
|
;* it under the terms of the GNU General Public License as published by
|
||||||
|
;* the Free Software Foundation; either version 2 of the License, or
|
||||||
|
;* any later version.
|
||||||
|
;*
|
||||||
|
;* This program is distributed in the hope that it will be useful,
|
||||||
|
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||||
|
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||||
|
;* GNU General Public License for more details.
|
||||||
|
;*
|
||||||
|
;* You should have received a copy of the GNU General Public License
|
||||||
|
;* along with this program; if not, write to the Free Software
|
||||||
|
;* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
||||||
|
;
|
||||||
|
;****************************************************************
|
||||||
|
;
|
||||||
|
; Glide64 - Glide Plugin for Nintendo 64 emulators
|
||||||
|
; Project started on December 29th, 2001
|
||||||
|
;
|
||||||
|
; Authors:
|
||||||
|
; Dave2001, original author, founded the project in 2001, left it in 2002
|
||||||
|
; Gugaman, joined the project in 2002, left it in 2002
|
||||||
|
; Sergey 'Gonetz' Lipski, joined the project in 2002, main author since fall of 2002
|
||||||
|
; Hiroshi 'KoolSmoky' Morii, joined the project in 2007
|
||||||
|
;
|
||||||
|
;****************************************************************
|
||||||
|
;
|
||||||
|
; To modify Glide64:
|
||||||
|
; * Write your name and (optional)email, commented by your work, so I know who did it, and so that you can find which parts you modified when it comes time to send it to me.
|
||||||
|
; * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
||||||
|
;
|
||||||
|
;****************************************************************
|
||||||
|
|
||||||
|
%include "inc/c32.mac"
|
||||||
|
|
||||||
|
segment .text
|
||||||
|
*/
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) DetectSIMD(int func, int * iedx, int * iecx)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
mov eax,[func]
|
||||||
|
cpuid
|
||||||
|
mov eax,[iedx]
|
||||||
|
mov [eax],edx
|
||||||
|
mov eax,[iecx]
|
||||||
|
mov [eax],ecx
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
/****************************************************************
|
||||||
|
;
|
||||||
|
; ******** SSE ********
|
||||||
|
;
|
||||||
|
;****************************************************************/
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) TransformVectorSSE(float *src, float *dst, float mat[4][4])
|
||||||
|
{
|
||||||
|
__asm
|
||||||
|
{
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
mov ecx,[src]
|
||||||
|
mov eax,[dst]
|
||||||
|
mov edx,[mat]
|
||||||
|
|
||||||
|
movss xmm0,[ecx] ; 0 0 0 src[0]
|
||||||
|
movss xmm5,[edx] ; 0 0 0 mat[0][0]
|
||||||
|
movhps xmm5,[edx+4] ; mat[0][2] mat[0][1] 0 mat[0][0]
|
||||||
|
shufps xmm0,xmm0, 0 ; src[0] src[0] src[0] src[0]
|
||||||
|
movss xmm1,[ecx+4] ; 0 0 0 src[1]
|
||||||
|
movss xmm3,[edx+16] ; 0 0 0 mat[1][0]
|
||||||
|
movhps xmm3,[edx+20] ; mat[1][2] mat[1][1] 0 mat[1][0]
|
||||||
|
shufps xmm1,xmm1, 0 ; src[1] src[1] src[1] src[1]
|
||||||
|
mulps xmm0,xmm5 ; mat[0][2]*src[0] mat[0][1]*src[0] 0 mat[0][0]*src[0]
|
||||||
|
mulps xmm1,xmm3 ; mat[1][2]*src[1] mat[1][1]*src[1] 0 mat[1][0]*src[1]
|
||||||
|
movss xmm2,[ecx+8] ; 0 0 0 src[2]
|
||||||
|
shufps xmm2,xmm2, 0 ; src[2] src[2] src[2] src[2]
|
||||||
|
movss xmm4,[edx+32] ; 0 0 0 mat[2][0]
|
||||||
|
movhps xmm4,[edx+36] ; mat[2][2] mat[2][1] 0 mat[2][0]
|
||||||
|
addps xmm0,xmm1 ; mat[0][2]*src[0]+mat[1][2]*src[1] mat[0][1]*src[0]+mat[1][1]*src[1] 0 mat[0][0]*src[0]+mat[1][0]*src[1]
|
||||||
|
mulps xmm2,xmm4 ; mat[2][2]*src[2] mat[2][1]*src[2] 0 mat[2][0]*src[2]
|
||||||
|
addps xmm0,xmm2 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2] mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2] 0 mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
||||||
|
movss [eax],xmm0 ; mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
||||||
|
movhps [eax+4],xmm0 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2] mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2]
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) MulMatricesSSE(float m1[4][4],float m2[4][4],float r[4][4])
|
||||||
|
{
|
||||||
|
__asm
|
||||||
|
{
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
mov eax,[r]
|
||||||
|
mov ecx,[m1]
|
||||||
|
mov edx,[m2]
|
||||||
|
|
||||||
|
movaps xmm0,[edx]
|
||||||
|
movaps xmm1,[edx+16]
|
||||||
|
movaps xmm2,[edx+32]
|
||||||
|
movaps xmm3,[edx+48]
|
||||||
|
|
||||||
|
; r[0][0],r[0][1],r[0][2],r[0][3]
|
||||||
|
|
||||||
|
movaps xmm4,[ecx]
|
||||||
|
movaps xmm5,xmm4
|
||||||
|
movaps xmm6,xmm4
|
||||||
|
movaps xmm7,xmm4
|
||||||
|
|
||||||
|
shufps xmm4,xmm4,00000000b
|
||||||
|
shufps xmm5,xmm5,01010101b
|
||||||
|
shufps xmm6,xmm6,10101010b
|
||||||
|
shufps xmm7,xmm7,11111111b
|
||||||
|
|
||||||
|
mulps xmm4,xmm0
|
||||||
|
mulps xmm5,xmm1
|
||||||
|
mulps xmm6,xmm2
|
||||||
|
mulps xmm7,xmm3
|
||||||
|
|
||||||
|
addps xmm4,xmm5
|
||||||
|
addps xmm4,xmm6
|
||||||
|
addps xmm4,xmm7
|
||||||
|
|
||||||
|
movaps [eax],xmm4
|
||||||
|
|
||||||
|
; r[1][0],r[1][1],r[1][2],r[1][3]
|
||||||
|
|
||||||
|
movaps xmm4,[ecx+16]
|
||||||
|
movaps xmm5,xmm4
|
||||||
|
movaps xmm6,xmm4
|
||||||
|
movaps xmm7,xmm4
|
||||||
|
|
||||||
|
shufps xmm4,xmm4,00000000b
|
||||||
|
shufps xmm5,xmm5,01010101b
|
||||||
|
shufps xmm6,xmm6,10101010b
|
||||||
|
shufps xmm7,xmm7,11111111b
|
||||||
|
|
||||||
|
mulps xmm4,xmm0
|
||||||
|
mulps xmm5,xmm1
|
||||||
|
mulps xmm6,xmm2
|
||||||
|
mulps xmm7,xmm3
|
||||||
|
|
||||||
|
addps xmm4,xmm5
|
||||||
|
addps xmm4,xmm6
|
||||||
|
addps xmm4,xmm7
|
||||||
|
|
||||||
|
movaps [eax+16],xmm4
|
||||||
|
|
||||||
|
|
||||||
|
; r[2][0],r[2][1],r[2][2],r[2][3]
|
||||||
|
|
||||||
|
movaps xmm4,[ecx+32]
|
||||||
|
movaps xmm5,xmm4
|
||||||
|
movaps xmm6,xmm4
|
||||||
|
movaps xmm7,xmm4
|
||||||
|
|
||||||
|
shufps xmm4,xmm4,00000000b
|
||||||
|
shufps xmm5,xmm5,01010101b
|
||||||
|
shufps xmm6,xmm6,10101010b
|
||||||
|
shufps xmm7,xmm7,11111111b
|
||||||
|
|
||||||
|
mulps xmm4,xmm0
|
||||||
|
mulps xmm5,xmm1
|
||||||
|
mulps xmm6,xmm2
|
||||||
|
mulps xmm7,xmm3
|
||||||
|
|
||||||
|
addps xmm4,xmm5
|
||||||
|
addps xmm4,xmm6
|
||||||
|
addps xmm4,xmm7
|
||||||
|
|
||||||
|
movaps [eax+32],xmm4
|
||||||
|
|
||||||
|
; r[3][0],r[3][1],r[3][2],r[3][3]
|
||||||
|
|
||||||
|
movaps xmm4,[ecx+48]
|
||||||
|
movaps xmm5,xmm4
|
||||||
|
movaps xmm6,xmm4
|
||||||
|
movaps xmm7,xmm4
|
||||||
|
|
||||||
|
shufps xmm4,xmm4,00000000b
|
||||||
|
shufps xmm5,xmm5,01010101b
|
||||||
|
shufps xmm6,xmm6,10101010b
|
||||||
|
shufps xmm7,xmm7,11111111b
|
||||||
|
|
||||||
|
mulps xmm4,xmm0
|
||||||
|
mulps xmm5,xmm1
|
||||||
|
mulps xmm6,xmm2
|
||||||
|
mulps xmm7,xmm3
|
||||||
|
|
||||||
|
addps xmm4,xmm5
|
||||||
|
addps xmm4,xmm6
|
||||||
|
addps xmm4,xmm7
|
||||||
|
|
||||||
|
movaps [eax+48],xmm4
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) NormalizeVectorSSE (float *v)
|
||||||
|
{
|
||||||
|
_asm
|
||||||
|
{
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
mov edx, [v]
|
||||||
|
movaps xmm0, [edx] ; x y z 0
|
||||||
|
movaps xmm2, xmm0 ; x y z 0
|
||||||
|
mulps xmm0, xmm0 ; x*x y*y z*z 0
|
||||||
|
movaps xmm1, xmm0 ; x*x y*y z*z 0
|
||||||
|
shufps xmm0, xmm1, 0x4e ; z*z 0 x*x y*y
|
||||||
|
addps xmm0, xmm1 ; x*x+z*z y*y z*z+x*x y*y
|
||||||
|
movaps xmm1, xmm0 ; x*x+z*z y*y z*z+x*x y*y
|
||||||
|
shufps xmm1, xmm1, 0x11 ; y*y z*z+x*x y*y z*z+x*x
|
||||||
|
addps xmm0, xmm1 ; x*x+z*z+y*y
|
||||||
|
rsqrtps xmm0, xmm0 ; 1.0/sqrt(x*x+z*z+y*y)
|
||||||
|
mulps xmm2, xmm0 ; x/sqrt(x*x+z*z+y*y) y/sqrt(x*x+z*z+y*y) z/sqrt(x*x+z*z+y*y) 0
|
||||||
|
movaps [edx], xmm2
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
/*****************************************************************
|
||||||
|
;
|
||||||
|
; ******** SSE3 ********
|
||||||
|
;
|
||||||
|
;****************************************************************/
|
||||||
|
|
||||||
|
float __declspec(naked) DotProductSSE3(register float *v1, register float *v2)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
mov eax,[v1]
|
||||||
|
mov edx,[v2]
|
||||||
|
movaps xmm0, [eax]
|
||||||
|
mulps xmm0, [edx]
|
||||||
|
haddps xmm0, xmm0
|
||||||
|
haddps xmm0, xmm0
|
||||||
|
; movss eax, xmm0
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
/****************************************************************
|
||||||
|
;
|
||||||
|
; ******** 3DNOW ********
|
||||||
|
;
|
||||||
|
;****************************************************************/
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) TransformVector3DNOW(float *src, float *dst, float mat[4][4])
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
femms
|
||||||
|
mov ecx,[src]
|
||||||
|
mov eax,[dst]
|
||||||
|
mov edx,[mat]
|
||||||
|
movq mm0,[ecx] ; src[1] src[0]
|
||||||
|
movd mm2,[ecx+8] ; 0 src[2]
|
||||||
|
movq mm1,mm0 ; src[1] src[0]
|
||||||
|
punpckldq mm0,mm0 ; src[0] src[0]
|
||||||
|
punpckhdq mm1,mm1 ; src[1] src[1]
|
||||||
|
punpckldq mm2,mm2 ; src[2] src[2]
|
||||||
|
movq mm3,mm0 ; src[0] src[0]
|
||||||
|
movq mm4,mm1 ; src[1] src[1]
|
||||||
|
movq mm5,mm2 ; src[2] src[2]
|
||||||
|
pfmul mm0,[edx] ; src[0]*mat[0][1] src[0]*mat[0][0]
|
||||||
|
pfmul mm3,[edx+8] ; 0 src[0]*mat[0][2]
|
||||||
|
pfmul mm1,[edx+16] ; src[1]*mat[1][1] src[1]*mat[1][0]
|
||||||
|
pfmul mm4,[edx+24] ; 0 src[1]*mat[1][2]
|
||||||
|
pfmul mm2,[edx+32] ; src[2]*mat[2][1] src[2]*mat[2][0]
|
||||||
|
pfmul mm5,[edx+40] ; 0 src[2]*mat[2][2]
|
||||||
|
pfadd mm0,mm1 ; src[0]*mat[0][1]+src[1]*mat[1][1] src[0]*mat[0][0]+src[1]*mat[1][0]
|
||||||
|
pfadd mm3,mm4 ; 0 src[0]*mat[0][2]+src[1]*mat[1][2]
|
||||||
|
pfadd mm0,mm2 ; src[0]*mat[0][1]+src[1]*mat[1][1]+src[2]*mat[2][1] src[0]*mat[0][0]+src[1]*mat[1][0]+src[2]*mat[2][0]
|
||||||
|
pfadd mm3,mm5 ; 0 src[0]*mat[0][2]+src[1]*mat[1][2]+src[2]*mat[2][2]
|
||||||
|
movq [eax],mm0 ; mat[0][1]*src[0]+mat[1][1]*src[1]+mat[2][1]*src[2] mat[0][0]*src[0]+mat[1][0]*src[1]+mat[2][0]*src[2]
|
||||||
|
movd [eax+8],mm3 ; mat[0][2]*src[0]+mat[1][2]*src[1]+mat[2][2]*src[2]
|
||||||
|
femms
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) InverseTransformVector3DNOW(float *src, float *dst, float mat[4][4])
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
femms
|
||||||
|
mov ecx,[src]
|
||||||
|
mov eax,[dst]
|
||||||
|
mov edx,[mat]
|
||||||
|
movq mm0,[ecx] ; src[1] src[0]
|
||||||
|
movd mm4,[ecx+8] ; 0 src[2]
|
||||||
|
movq mm1,mm0 ; src[1] src[0]
|
||||||
|
pfmul mm0,[edx] ; src[1]*mat[0][1] src[0]*mat[0][0]
|
||||||
|
movq mm5,mm4 ; 0 src[2]
|
||||||
|
pfmul mm4,[edx+8] ; 0 src[2]*mat[0][2]
|
||||||
|
movq mm2,mm1 ; src[1] src[0]
|
||||||
|
pfmul mm1,[edx+16] ; src[1]*mat[1][1] src[0]*mat[1][0]
|
||||||
|
movq mm6,mm5 ; 0 src[2]
|
||||||
|
pfmul mm5,[edx+24] ; 0 src[2]*mat[1][2]
|
||||||
|
movq mm3,mm2 ; src[1] src[0]
|
||||||
|
pfmul mm2,[edx+32] ; src[1]*mat[2][1] src[0]*mat[2][0]
|
||||||
|
movq mm7,mm6 ; 0 src[2]
|
||||||
|
pfmul mm6,[edx+40] ; 0 src[2]*mat[2][2]
|
||||||
|
pfacc mm0,mm4 ; src[2]*mat[0][2] src[1]*mat[0][1]+src[0]*mat[0][0]
|
||||||
|
pfacc mm1,mm5 ; src[2]*mat[1][2] src[1]*mat[1][1]+src[0]*mat[1][0]
|
||||||
|
pfacc mm2,mm6 ; src[2]*mat[2][2] src[1]*mat[2][1]+src[0]*mat[2][0]
|
||||||
|
pfacc mm0,mm1 ; src[2]*mat[1][2]+src[1]*mat[1][1]+src[0]*mat[1][0] src[2]*mat[0][2]+src[1]*mat[0][1]+src[0]*mat[0][0]
|
||||||
|
pfacc mm2,mm3 ; 0 src[2]*mat[2][2]+src[1]*mat[2][1]+src[0]*mat[2][0]
|
||||||
|
movq [eax],mm0 ; mat[1][0]*src[0]+mat[1][1]*src[1]+mat[1][2]*src[2] mat[0][0]*src[0]+mat[0][1]*src[1]+mat[0][2]*src[2]
|
||||||
|
movd [eax+8],mm2 ; mat[2][0]*src[0]+mat[2][1]*src[1]+mat[2][2]*src[2]
|
||||||
|
femms
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) MulMatrices3DNOW(float m1[4][4],float m2[4][4],float r[4][4])
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
femms
|
||||||
|
mov ecx,[m1]
|
||||||
|
mov eax,[r]
|
||||||
|
mov edx,[m2]
|
||||||
|
|
||||||
|
movq mm0,[ecx]
|
||||||
|
movq mm1,[ecx+8]
|
||||||
|
movq mm4,[edx]
|
||||||
|
punpckhdq mm2,mm0
|
||||||
|
movq mm5,[edx+16]
|
||||||
|
punpckhdq mm3,mm1
|
||||||
|
movq mm6,[edx+32]
|
||||||
|
punpckldq mm0,mm0
|
||||||
|
punpckldq mm1,mm1
|
||||||
|
pfmul mm4,mm0
|
||||||
|
punpckhdq mm2,mm2
|
||||||
|
pfmul mm0,[edx+8]
|
||||||
|
movq mm7,[edx+48]
|
||||||
|
pfmul mm5,mm2
|
||||||
|
punpckhdq mm3,mm3
|
||||||
|
pfmul mm2,[edx+24]
|
||||||
|
pfmul mm6,mm1
|
||||||
|
pfadd mm5,mm4
|
||||||
|
pfmul mm1,[edx+40]
|
||||||
|
pfadd mm2,mm0
|
||||||
|
pfmul mm7,mm3
|
||||||
|
pfadd mm6,mm5
|
||||||
|
pfmul mm3,[edx+56]
|
||||||
|
pfadd mm2,mm1
|
||||||
|
pfadd mm7,mm6
|
||||||
|
movq mm0,[ecx+16]
|
||||||
|
pfadd mm3,mm2
|
||||||
|
movq mm1,[ecx+24]
|
||||||
|
movq [eax],mm7
|
||||||
|
movq mm4,[edx]
|
||||||
|
movq [eax+8],mm3
|
||||||
|
|
||||||
|
punpckhdq mm2,mm0
|
||||||
|
movq mm5,[edx+16]
|
||||||
|
punpckhdq mm3,mm1
|
||||||
|
movq mm6,[edx+32]
|
||||||
|
punpckldq mm0,mm0
|
||||||
|
punpckldq mm1,mm1
|
||||||
|
pfmul mm4,mm0
|
||||||
|
punpckhdq mm2,mm2
|
||||||
|
pfmul mm0,[edx+8]
|
||||||
|
movq mm7,[edx+48]
|
||||||
|
pfmul mm5,mm2
|
||||||
|
punpckhdq mm3,mm3
|
||||||
|
pfmul mm2,[edx+24]
|
||||||
|
pfmul mm6,mm1
|
||||||
|
pfadd mm5,mm4
|
||||||
|
pfmul mm1,[edx+40]
|
||||||
|
pfadd mm2,mm0
|
||||||
|
pfmul mm7,mm3
|
||||||
|
pfadd mm6,mm5
|
||||||
|
pfmul mm3,[edx+56]
|
||||||
|
pfadd mm2,mm1
|
||||||
|
pfadd mm7,mm6
|
||||||
|
movq mm0,[ecx+32]
|
||||||
|
pfadd mm3,mm2
|
||||||
|
movq mm1,[ecx+40]
|
||||||
|
movq [eax+16],mm7
|
||||||
|
movq mm4,[edx]
|
||||||
|
movq [eax+24],mm3
|
||||||
|
|
||||||
|
punpckhdq mm2,mm0
|
||||||
|
movq mm5,[edx+16]
|
||||||
|
punpckhdq mm3,mm1
|
||||||
|
movq mm6,[edx+32]
|
||||||
|
punpckldq mm0,mm0
|
||||||
|
punpckldq mm1,mm1
|
||||||
|
pfmul mm4,mm0
|
||||||
|
punpckhdq mm2,mm2
|
||||||
|
pfmul mm0,[edx+8]
|
||||||
|
movq mm7,[edx+48]
|
||||||
|
pfmul mm5,mm2
|
||||||
|
punpckhdq mm3,mm3
|
||||||
|
pfmul mm2,[edx+24]
|
||||||
|
pfmul mm6,mm1
|
||||||
|
pfadd mm5,mm4
|
||||||
|
pfmul mm1,[edx+40]
|
||||||
|
pfadd mm2,mm0
|
||||||
|
pfmul mm7,mm3
|
||||||
|
pfadd mm6,mm5
|
||||||
|
pfmul mm3,[edx+56]
|
||||||
|
pfadd mm2,mm1
|
||||||
|
pfadd mm7,mm6
|
||||||
|
movq mm0,[ecx+48]
|
||||||
|
pfadd mm3,mm2
|
||||||
|
movq mm1,[ecx+56]
|
||||||
|
movq [eax+32],mm7
|
||||||
|
movq mm4,[edx]
|
||||||
|
movq [eax+40],mm3
|
||||||
|
|
||||||
|
punpckhdq mm2,mm0
|
||||||
|
movq mm5,[edx+16]
|
||||||
|
punpckhdq mm3,mm1
|
||||||
|
movq mm6,[edx+32]
|
||||||
|
punpckldq mm0,mm0
|
||||||
|
punpckldq mm1,mm1
|
||||||
|
pfmul mm4,mm0
|
||||||
|
punpckhdq mm2,mm2
|
||||||
|
pfmul mm0,[edx+8]
|
||||||
|
movq mm7,[edx+48]
|
||||||
|
pfmul mm5,mm2
|
||||||
|
punpckhdq mm3,mm3
|
||||||
|
pfmul mm2,[edx+24]
|
||||||
|
pfmul mm6,mm1
|
||||||
|
pfadd mm5,mm4
|
||||||
|
pfmul mm1,[edx+40]
|
||||||
|
pfadd mm2,mm0
|
||||||
|
pfmul mm7,mm3
|
||||||
|
pfadd mm6,mm5
|
||||||
|
pfmul mm3,[edx+56]
|
||||||
|
pfadd mm2,mm1
|
||||||
|
pfadd mm7,mm6
|
||||||
|
pfadd mm3,mm2
|
||||||
|
movq [eax+48],mm7
|
||||||
|
movq [eax+56],mm3
|
||||||
|
femms
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" float __declspec(naked) DotProduct3DNOW(register float *v1, register float *v2)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
|
||||||
|
femms
|
||||||
|
mov edx,[v1]
|
||||||
|
mov eax,[v2]
|
||||||
|
movq mm0,[edx]
|
||||||
|
movq mm3,[eax]
|
||||||
|
pfmul mm0,mm3
|
||||||
|
movq mm2,[edx+8]
|
||||||
|
movq mm1,[eax+8]
|
||||||
|
pfacc mm0,mm0
|
||||||
|
pfmul mm1,mm2
|
||||||
|
pfadd mm0,mm1
|
||||||
|
movd eax,mm0
|
||||||
|
femms
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
extern "C" void __declspec(naked) NormalizeVector3DNOW(float *v)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
femms
|
||||||
|
mov edx,[v]
|
||||||
|
movq mm0,[edx]
|
||||||
|
movq mm3,[edx+8]
|
||||||
|
movq mm1,mm0
|
||||||
|
movq mm2,mm3
|
||||||
|
pfmul mm0,mm0
|
||||||
|
pfmul mm3,mm3
|
||||||
|
pfacc mm0,mm0
|
||||||
|
pfadd mm0,mm3
|
||||||
|
;movq mm4,mm0 ; prepare for 24bit precision
|
||||||
|
;punpckldq mm4,mm4 ; prepare for 24bit precision
|
||||||
|
pfrsqrt mm0,mm0 ; 15bit precision 1/sqrtf(v)
|
||||||
|
;movq mm3,mm0
|
||||||
|
;pfmul mm0,mm0
|
||||||
|
;pfrsqit1 mm0,mm4
|
||||||
|
;pfrcpit2 mm0,mm3 ; 24bit precision 1/sqrtf(v)
|
||||||
|
pfmul mm1,mm0
|
||||||
|
pfmul mm2,mm0
|
||||||
|
movq [edx],mm1
|
||||||
|
movq [edx+8],mm2
|
||||||
|
femms
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
@ -44,8 +44,6 @@
|
||||||
|
|
||||||
// -*- C++ -*- generated by wxGlade 0.6.3 on Wed Oct 08 18:56:23 2008
|
// -*- C++ -*- generated by wxGlade 0.6.3 on Wed Oct 08 18:56:23 2008
|
||||||
|
|
||||||
#define wxNO_QA_LIB
|
|
||||||
|
|
||||||
#include <wx/wx.h>
|
#include <wx/wx.h>
|
||||||
#include <wx/image.h>
|
#include <wx/image.h>
|
||||||
// begin wxGlade: ::dependencies
|
// begin wxGlade: ::dependencies
|
||||||
|
|
|
||||||
|
|
@ -85,37 +85,9 @@ static int right_height, left_height;
|
||||||
static int right_x, right_dxdy, left_x, left_dxdy;
|
static int right_x, right_dxdy, left_x, left_dxdy;
|
||||||
static int left_z, left_dzdy;
|
static int left_z, left_dzdy;
|
||||||
|
|
||||||
int imul16(int x, int y)
|
extern "C" int imul16(int x, int y);
|
||||||
{
|
extern "C" int imul14(int x, int y);
|
||||||
_asm {
|
extern "C" int idiv16(int x, int y);
|
||||||
mov eax, [x]
|
|
||||||
mov edx, [y]
|
|
||||||
imul edx
|
|
||||||
shrd eax,edx,16
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
int imul14(int x, int y)
|
|
||||||
{
|
|
||||||
_asm {
|
|
||||||
mov eax, [x]
|
|
||||||
mov edx, [y]
|
|
||||||
imul edx
|
|
||||||
shrd eax,edx,14
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
int idiv16(int x, int y)
|
|
||||||
{
|
|
||||||
_asm {
|
|
||||||
mov eax, [x]
|
|
||||||
mov ebx, [y]
|
|
||||||
mov edx,eax
|
|
||||||
sar edx,16
|
|
||||||
shl eax,16
|
|
||||||
idiv ebx
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
__inline int iceil(int x)
|
__inline int iceil(int x)
|
||||||
{
|
{
|
||||||
|
|
|
||||||
|
|
@ -1,51 +0,0 @@
|
||||||
;/*
|
|
||||||
;* Glide64 - Glide video plugin for Nintendo 64 emulators.
|
|
||||||
;*
|
|
||||||
;* This program is free software; you can redistribute it and/or modify
|
|
||||||
;* it under the terms of the GNU General Public License as published by
|
|
||||||
;* the Free Software Foundation; either version 2 of the License, or
|
|
||||||
;* any later version.
|
|
||||||
;*
|
|
||||||
;* This program is distributed in the hope that it will be useful,
|
|
||||||
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
||||||
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
||||||
;* GNU General Public License for more details.
|
|
||||||
;*
|
|
||||||
;* You should have received a copy of the GNU General Public License
|
|
||||||
;* along with this program; if not, write to the Free Software
|
|
||||||
;* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
|
||||||
;*/
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; Glide64 - Glide Plugin for Nintendo 64 emulators
|
|
||||||
; Project started on December 29th, 2001
|
|
||||||
;
|
|
||||||
; Authors:
|
|
||||||
; Dave2001, original author, founded the project in 2001, left it in 2002
|
|
||||||
; Gugaman, joined the project in 2002, left it in 2002
|
|
||||||
; Sergey 'Gonetz' Lipski, joined the project in 2002, main author since fall of 2002
|
|
||||||
; Hiroshi 'KoolSmoky' Morii, joined the project in 2007
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
;
|
|
||||||
; To modify Glide64:
|
|
||||||
; * Write your name and (optional)email, commented by your work, so I know who did it, and so that you can find which parts you modified when it comes time to send it to me.
|
|
||||||
; * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
|
||||||
;
|
|
||||||
;****************************************************************
|
|
||||||
|
|
||||||
%include "inc/c32.mac"
|
|
||||||
|
|
||||||
segment .text
|
|
||||||
|
|
||||||
endproc ;imul14
|
|
||||||
|
|
||||||
;(x << 16) / y
|
|
||||||
proc idiv16
|
|
||||||
CPU 586
|
|
||||||
|
|
||||||
%$x arg
|
|
||||||
%$y arg
|
|
||||||
|
|
||||||
endproc ;idiv16
|
|
||||||
|
|
@ -0,0 +1,83 @@
|
||||||
|
/*
|
||||||
|
* Glide64 - Glide video plugin for Nintendo 64 emulators.
|
||||||
|
*
|
||||||
|
* This program is free software; you can redistribute it and/or modify
|
||||||
|
* it under the terms of the GNU General Public License as published by
|
||||||
|
* the Free Software Foundation; either version 2 of the License, or
|
||||||
|
* any later version.
|
||||||
|
*
|
||||||
|
* This program is distributed in the hope that it will be useful,
|
||||||
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||||
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||||
|
* GNU General Public License for more details.
|
||||||
|
*
|
||||||
|
* You should have received a copy of the GNU General Public License
|
||||||
|
* along with this program; if not, write to the Free Software
|
||||||
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
||||||
|
*/
|
||||||
|
|
||||||
|
/****************************************************************
|
||||||
|
;
|
||||||
|
; Glide64 - Glide Plugin for Nintendo 64 emulators
|
||||||
|
; Project started on December 29th, 2001
|
||||||
|
;
|
||||||
|
; Authors:
|
||||||
|
; Dave2001, original author, founded the project in 2001, left it in 2002
|
||||||
|
; Gugaman, joined the project in 2002, left it in 2002
|
||||||
|
; Sergey 'Gonetz' Lipski, joined the project in 2002, main author since fall of 2002
|
||||||
|
; Hiroshi 'KoolSmoky' Morii, joined the project in 2007
|
||||||
|
;
|
||||||
|
;****************************************************************
|
||||||
|
;
|
||||||
|
; To modify Glide64:
|
||||||
|
; * Write your name and (optional)email, commented by your work, so I know who did it, and so that you can find which parts you modified when it comes time to send it to me.
|
||||||
|
; * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
||||||
|
;
|
||||||
|
;****************************************************************/
|
||||||
|
|
||||||
|
// (x * y) >> 16
|
||||||
|
extern "C" int __declspec(naked) imul16(int x, int y)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
mov eax, [x]
|
||||||
|
mov edx, [y]
|
||||||
|
imul edx
|
||||||
|
shrd eax,edx,16
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
//(x * y) >> 14
|
||||||
|
extern "C" int __declspec(naked) imul14(int x, int y)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
mov eax, [x]
|
||||||
|
mov edx, [y]
|
||||||
|
imul edx
|
||||||
|
shrd eax,edx,14
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
//(x << 16) / y
|
||||||
|
extern "C" int __declspec(naked) idiv16(int x, int y)
|
||||||
|
{
|
||||||
|
_asm {
|
||||||
|
push ebp
|
||||||
|
mov ebp,esp
|
||||||
|
mov eax, [x]
|
||||||
|
mov ebx, [y]
|
||||||
|
mov edx,eax
|
||||||
|
sar edx,16
|
||||||
|
shl eax,16
|
||||||
|
idiv ebx
|
||||||
|
leave
|
||||||
|
ret
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
@ -345,7 +345,7 @@
|
||||||
>
|
>
|
||||||
</File>
|
</File>
|
||||||
<File
|
<File
|
||||||
RelativePath="3dmathSIMD.asm"
|
RelativePath=".\3dmathSIMD.asm.cpp"
|
||||||
>
|
>
|
||||||
</File>
|
</File>
|
||||||
<File
|
<File
|
||||||
|
|
@ -389,7 +389,7 @@
|
||||||
>
|
>
|
||||||
</File>
|
</File>
|
||||||
<File
|
<File
|
||||||
RelativePath="FixedPoint.asm"
|
RelativePath=".\FixedPoint.asm.cpp"
|
||||||
>
|
>
|
||||||
</File>
|
</File>
|
||||||
<File
|
<File
|
||||||
|
|
|
||||||
|
|
@ -36,6 +36,7 @@
|
||||||
// * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
// * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
#include "Gfx #1.3.h"
|
#include "Gfx #1.3.h"
|
||||||
#include <wx/fileconf.h>
|
#include <wx/fileconf.h>
|
||||||
#include <wx/wfstream.h>
|
#include <wx/wfstream.h>
|
||||||
|
|
|
||||||
|
|
@ -37,9 +37,9 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmMirror16bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
extern "C" void asmMirror16bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
||||||
void asmWrap16bS (int tex, int start, int height, int mask, int line, int full, int count);
|
extern "C" void asmWrap16bS (int tex, int start, int height, int mask, int line, int full, int count);
|
||||||
void asmClamp16bS (int tex, int constant, int height,int line, int full, int count);
|
extern "C" void asmClamp16bS (int tex, int constant, int height,int line, int full, int count);
|
||||||
|
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// 16-bit Horizontal Mirror
|
// 16-bit Horizontal Mirror
|
||||||
|
|
|
||||||
|
|
@ -41,9 +41,9 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmMirror32bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
extern "C" void asmMirror32bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
||||||
void asmWrap32bS (int tex, int start, int height, int mask, int line, int full, int count);
|
extern "C" void asmWrap32bS (int tex, int start, int height, int mask, int line, int full, int count);
|
||||||
void asmClamp32bS (int tex, int constant, int height,int line, int full, int count);
|
extern "C" void asmClamp32bS (int tex, int constant, int height,int line, int full, int count);
|
||||||
|
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// 32-bit Horizontal Mirror
|
// 32-bit Horizontal Mirror
|
||||||
|
|
|
||||||
|
|
@ -40,9 +40,9 @@
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// 8-bit Horizontal Mirror
|
// 8-bit Horizontal Mirror
|
||||||
|
|
||||||
void asmMirror8bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
extern "C" void asmMirror8bS (int tex, int start, int width, int height, int mask, int line, int full, int count);
|
||||||
void asmWrap8bS (int tex, int start, int height, int mask, int line, int full, int count);
|
extern "C" void asmWrap8bS (int tex, int start, int height, int mask, int line, int full, int count);
|
||||||
void asmClamp8bS (int tex, int constant, int height,int line, int full, int count);
|
extern "C" void asmClamp8bS (int tex, int constant, int height,int line, int full, int count);
|
||||||
|
|
||||||
void Mirror8bS (wxUint32 tex, wxUint32 mask, wxUint32 max_width, wxUint32 real_width, wxUint32 height)
|
void Mirror8bS (wxUint32 tex, wxUint32 mask, wxUint32 max_width, wxUint32 real_width, wxUint32 height)
|
||||||
{
|
{
|
||||||
|
|
|
||||||
|
|
@ -37,10 +37,10 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmTexConv_ARGB1555_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
extern "C" void asmTexConv_ARGB1555_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
||||||
void asmTexConv_AI88_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
extern "C" void asmTexConv_AI88_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
||||||
void asmTexConv_AI44_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
extern "C" void asmTexConv_AI44_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
||||||
void asmTexConv_A8_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
extern "C" void asmTexConv_A8_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int size);
|
||||||
|
|
||||||
void TexConv_ARGB1555_ARGB4444 (wxUIntPtr src, wxUIntPtr dst, int width, int height)
|
void TexConv_ARGB1555_ARGB4444 (wxUIntPtr src, wxUIntPtr dst, int width, int height)
|
||||||
{
|
{
|
||||||
|
|
|
||||||
|
|
@ -37,15 +37,15 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmLoad16bRGBA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
extern "C" void asmLoad16bRGBA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
||||||
void asmLoad16bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
extern "C" void asmLoad16bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
||||||
|
|
||||||
|
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// Size: 2, Format: 0
|
// Size: 2, Format: 0
|
||||||
//
|
//
|
||||||
|
|
||||||
wxUint32 Load16bRGBA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int tile)
|
wxUint32 Load16bRGBA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int /*tile*/)
|
||||||
{
|
{
|
||||||
if (wid_64 < 1) wid_64 = 1;
|
if (wid_64 < 1) wid_64 = 1;
|
||||||
if (height < 1) height = 1;
|
if (height < 1) height = 1;
|
||||||
|
|
@ -60,7 +60,7 @@ wxUint32 Load16bRGBA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int
|
||||||
// Size: 2, Format: 3
|
// Size: 2, Format: 3
|
||||||
//
|
//
|
||||||
|
|
||||||
wxUint32 Load16bIA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int tile)
|
wxUint32 Load16bIA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int /*tile*/)
|
||||||
{
|
{
|
||||||
if (wid_64 < 1) wid_64 = 1;
|
if (wid_64 < 1) wid_64 = 1;
|
||||||
if (height < 1) height = 1;
|
if (height < 1) height = 1;
|
||||||
|
|
@ -115,7 +115,7 @@ wxUint16 yuv_to_rgb565(wxUint8 y, wxUint8 u, wxUint8 v)
|
||||||
// Size: 2, Format: 1
|
// Size: 2, Format: 1
|
||||||
//
|
//
|
||||||
|
|
||||||
wxUint32 Load16bYUV (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int tile)
|
wxUint32 Load16bYUV (wxUIntPtr dst, wxUIntPtr src, int wid_64, int /*height*/, int /*line*/, int /*real_width*/, int tile)
|
||||||
{
|
{
|
||||||
wxUint32 * mb = (wxUint32*)(gfx.RDRAM+rdp.addr[rdp.tiles[tile].t_mem]); //pointer to the macro block
|
wxUint32 * mb = (wxUint32*)(gfx.RDRAM+rdp.addr[rdp.tiles[tile].t_mem]); //pointer to the macro block
|
||||||
wxUint16 * tex = (wxUint16*)dst;
|
wxUint16 * tex = (wxUint16*)dst;
|
||||||
|
|
|
||||||
|
|
@ -37,10 +37,10 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmLoad4bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
extern "C" void asmLoad4bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
||||||
void asmLoad4bIAPal (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
extern "C" void asmLoad4bIAPal (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
||||||
void asmLoad4bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
extern "C" void asmLoad4bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
||||||
void asmLoad4bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext);
|
extern "C" void asmLoad4bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext);
|
||||||
|
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// Size: 0, Format: 2
|
// Size: 0, Format: 2
|
||||||
|
|
|
||||||
|
|
@ -37,16 +37,16 @@
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void asmLoad8bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
extern "C" void asmLoad8bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
||||||
void asmLoad8bIA8 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
extern "C" void asmLoad8bIA8 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal);
|
||||||
void asmLoad8bIA4 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
extern "C" void asmLoad8bIA4 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext);
|
||||||
void asmLoad8bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext);
|
extern "C" void asmLoad8bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext);
|
||||||
|
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
// Size: 1, Format: 2
|
// Size: 1, Format: 2
|
||||||
//
|
//
|
||||||
|
|
||||||
wxUint32 Load8bCI (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int tile)
|
wxUint32 Load8bCI (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line, int real_width, int /*tile*/)
|
||||||
{
|
{
|
||||||
if (wid_64 < 1) wid_64 = 1;
|
if (wid_64 < 1) wid_64 = 1;
|
||||||
if (height < 1) height = 1;
|
if (height < 1) height = 1;
|
||||||
|
|
|
||||||
|
|
@ -54,7 +54,7 @@
|
||||||
Size: 0, Format: 2
|
Size: 0, Format: 2
|
||||||
2009 ported to NASM - Sergey (Gonetz) Lipski
|
2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
*****************************************************************/
|
*****************************************************************/
|
||||||
void __declspec(naked) asmLoad4bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
extern "C" void __declspec(naked) asmLoad4bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -430,7 +430,7 @@ end_y_loop:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmLoad4bIAPal (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
extern "C" void __declspec(naked) asmLoad4bIAPal (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -812,7 +812,7 @@ end_y_loop:
|
||||||
** BY GUGAMAN **
|
** BY GUGAMAN **
|
||||||
2009 ported to NASM - Sergey (Gonetz) Lipski
|
2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
*****************************************************************/
|
*****************************************************************/
|
||||||
void __declspec(naked) asmLoad4bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad4bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -1623,7 +1623,7 @@ end_y_loop:
|
||||||
// Size: 0, Format: 4
|
// Size: 0, Format: 4
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad4bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad4bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -1966,7 +1966,7 @@ end_y_loop:
|
||||||
// 2008.03.29 cleaned up - H.Morii
|
// 2008.03.29 cleaned up - H.Morii
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad8bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
extern "C" void __declspec(naked) asmLoad8bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -2198,7 +2198,7 @@ end_y_loop:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmLoad8bIA8 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
extern "C" void __declspec(naked) asmLoad8bIA8 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -2423,7 +2423,7 @@ end_y_loop:
|
||||||
// 2008.03.29 cleaned up - H.Morii
|
// 2008.03.29 cleaned up - H.Morii
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad8bIA4 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad8bIA4 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -2531,7 +2531,7 @@ end_y_loop:
|
||||||
// ** by Gugaman **
|
// ** by Gugaman **
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad8bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad8bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
push ebp
|
push ebp
|
||||||
|
|
@ -2616,7 +2616,7 @@ end_y_loop:
|
||||||
// 2008.03.29 cleaned up - H.Morii
|
// 2008.03.29 cleaned up - H.Morii
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad16bRGBA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad16bRGBA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -2720,7 +2720,7 @@ end_y_loop:
|
||||||
// 2008.03.29 cleaned up - H.Morii
|
// 2008.03.29 cleaned up - H.Morii
|
||||||
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
// 2009 ported to NASM - Sergey (Gonetz) Lipski
|
||||||
|
|
||||||
void __declspec(naked) asmLoad16bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
extern "C" void __declspec(naked) asmLoad16bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
ALIGN 4
|
ALIGN 4
|
||||||
|
|
@ -2795,7 +2795,7 @@ end_y_loop:
|
||||||
//8b textures mirror/clamp/wrap
|
//8b textures mirror/clamp/wrap
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void __declspec(naked) asmMirror8bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmMirror8bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm{
|
_asm{
|
||||||
ALIGN 4
|
ALIGN 4
|
||||||
|
|
@ -2856,7 +2856,7 @@ end_mirror_check:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmWrap8bS (int tex, int start, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmWrap8bS (int tex, int start, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -2903,7 +2903,7 @@ loop_x:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmClamp8bS (int tex, int constant, int height,int line, int full, int count)
|
extern "C" void __declspec(naked) asmClamp8bS (int tex, int constant, int height,int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -2949,7 +2949,7 @@ x_loop:
|
||||||
//16b textures mirror/clamp/wrap
|
//16b textures mirror/clamp/wrap
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void __declspec(naked) asmMirror16bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmMirror16bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3011,7 +3011,7 @@ end_mirror_check:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmWrap16bS (int tex, int start, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmWrap16bS (int tex, int start, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3058,7 +3058,7 @@ loop_x:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmClamp16bS (int tex, int constant, int height,int line, int full, int count)
|
extern "C" void __declspec(naked) asmClamp16bS (int tex, int constant, int height,int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3104,7 +3104,7 @@ x_loop:
|
||||||
//32b textures mirror/clamp/wrap
|
//32b textures mirror/clamp/wrap
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void __declspec(naked) asmMirror32bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmMirror32bS (int tex, int start, int width, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3166,7 +3166,7 @@ end_mirror_check:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmWrap32bS (int tex, int start, int height, int mask, int line, int full, int count)
|
extern "C" void __declspec(naked) asmWrap32bS (int tex, int start, int height, int mask, int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3213,7 +3213,7 @@ loop_x:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmClamp32bS (int tex, int constant, int height,int line, int full, int count)
|
extern "C" void __declspec(naked) asmClamp32bS (int tex, int constant, int height,int line, int full, int count)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3261,7 +3261,7 @@ x_loop:
|
||||||
//
|
//
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
|
|
||||||
void __declspec(naked) asmTexConv_ARGB1555_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
extern "C" void __declspec(naked) asmTexConv_ARGB1555_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3320,7 +3320,7 @@ tc1_loop:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmTexConv_AI88_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
extern "C" void __declspec(naked) asmTexConv_AI88_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3367,7 +3367,7 @@ tc1_loop:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmTexConv_AI44_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
extern "C" void __declspec(naked) asmTexConv_AI44_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3445,7 +3445,7 @@ tc1_loop:
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
void __declspec(naked) asmTexConv_A8_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
extern "C" void __declspec(naked) asmTexConv_A8_ARGB4444(wxUIntPtr src, wxUIntPtr dst, int isize)
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
@ -3541,7 +3541,7 @@ tc1_loop:
|
||||||
// esi = base_addr (preserved)
|
// esi = base_addr (preserved)
|
||||||
// edx = offset (preserved)
|
// edx = offset (preserved)
|
||||||
//****************************************************************
|
//****************************************************************
|
||||||
__declspec(naked) void CopyBlock ( void )
|
void __declspec(naked) CopyBlock ( void )
|
||||||
{
|
{
|
||||||
_asm {
|
_asm {
|
||||||
align 4
|
align 4
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue