ShuriZma
/
project64
mirror of https://github.com/project64/project64.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[Glide64] Sync texture code

This commit is contained in:
zilmar 2015-10-21 07:49:29 +11:00
parent 11ebb4e761
commit 57cf720392
3 changed files with 1001 additions and 2668 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

276
Source/Glide64/TexLoad16b.h
View File

@ -37,164 +37,130 @@
// //
//**************************************************************** //****************************************************************
extern "C" void __declspec(naked) asmLoad16bRGBA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext) static inline void load16bRGBA(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext)
{ {
_asm { uint32_t *v6;
align 4 uint32_t *v7;
push ebp int v8;
mov ebp,esp int v9;
push ebx uint32_t v10;
push esi uint32_t v11;
push edi uint32_t *v12;
uint32_t *v13;
int v14;
uint32_t v15;
uint32_t v16;
int v17;
int v18;
mov esi,[src] v6 = (uint32_t *)src;
mov edi,[dst] v7 = (uint32_t *)dst;
mov ecx,[height] v8 = height;
y_loop: do
push ecx
mov ecx,[wid_64]
x_loop:
mov eax,[esi] // read both pixels
mov ebx,[esi+4] // read both pixels
bswap eax
bswap ebx
ror ax,1
ror bx,1
ror eax,16
ror ebx,16
ror ax,1
ror bx,1
mov [edi],eax
mov [edi+4],ebx
add esi,8
add edi,8
dec ecx
jnz x_loop
pop ecx
dec ecx
jz end_y_loop
push ecx
mov eax,esi
add eax,[line]
mov esi,[src]
sub eax, esi
and eax, 0xFFF
add esi, eax
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
mov eax,[esi+4] // read both pixels
mov ebx,[esi] // read both pixels
bswap eax
bswap ebx
ror ax,1
ror bx,1
ror eax,16
ror ebx,16
ror ax,1
ror bx,1
mov [edi],eax
mov [edi+4],ebx
add esi,8
add edi,8
dec ecx
jnz x_loop_2
mov eax,esi
add eax,[line]
mov esi,[src]
sub eax, esi
and eax, 0xFFF
add esi, eax
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
}
}
extern "C" void __declspec(naked) asmLoad16bIA (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
{ {
_asm { v17 = v8;
ALIGN 4 v9 = wid_64;
do
push ebp {
mov ebp, esp v10 = bswap32(*v6);
push ebx v11 = bswap32(v6[1]);
push esi ALOWORD(v10) = __ROR__((uint16_t)v10, 1);
push edi ALOWORD(v11) = __ROR__((uint16_t)v11, 1);
v10 = __ROR__(v10, 16);
mov esi,[src] v11 = __ROR__(v11, 16);
mov edi,[dst] ALOWORD(v10) = __ROR__((uint16_t)v10, 1);
mov ecx,[height] ALOWORD(v11) = __ROR__((uint16_t)v11, 1);
y_loop: *v7 = v10;
push ecx v7[1] = v11;
mov ecx,[wid_64] v6 += 2;
x_loop: v7 += 2;
mov eax,[esi] // read both pixels --v9;
mov ebx,[esi+4] // read both pixels
mov [edi],eax
mov [edi+4],ebx
add esi,8
add edi,8
dec ecx
jnz x_loop
pop ecx
dec ecx
jz end_y_loop
push ecx
add esi,[line]
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
mov eax,[esi+4] // read both pixels
mov ebx,[esi] // read both pixels
mov [edi],eax
mov [edi+4],ebx
add esi,8
add edi,8
dec ecx
jnz x_loop_2
add esi,[line]
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
} }
while ( v9 );
if ( v17 == 1 )
break;
v18 = v17 - 1;
v12 = (uint32_t *)&src[(line + (uintptr_t)v6 - (uintptr_t)src) & 0xFFF];
v13 = (uint32_t *)((char *)v7 + ext);
v14 = wid_64;
do
{
v15 = bswap32(v12[1]);
v16 = bswap32(*v12);
ALOWORD(v15) = __ROR__((uint16_t)v15, 1);
ALOWORD(v16) = __ROR__((uint16_t)v16, 1);
v15 = __ROR__(v15, 16);
v16 = __ROR__(v16, 16);
ALOWORD(v15) = __ROR__((uint16_t)v15, 1);
ALOWORD(v16) = __ROR__((uint16_t)v16, 1);
*v13 = v15;
v13[1] = v16;
v12 += 2;
v13 += 2;
--v14;
} }
while ( v14 );
v6 = (uint32_t *)&src[(line + (uintptr_t)v12 - (uintptr_t)src) & 0xFFF];
v7 = (uint32_t *)((char *)v13 + ext);
v8 = v18 - 1;
}
while ( v18 != 1 );
}
static inline void load16bIA(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext)
{
uint32_t *v6;
uint32_t *v7;
int v8;
int v9;
uint32_t v10;
uint32_t *v11;
uint32_t *v12;
int v13;
uint32_t v14;
int v15;
int v16;
v6 = (uint32_t *)src;
v7 = (uint32_t *)dst;
v8 = height;
do
{
v15 = v8;
v9 = wid_64;
do
{
v10 = v6[1];
*v7 = *v6;
v7[1] = v10;
v6 += 2;
v7 += 2;
--v9;
}
while ( v9 );
if ( v15 == 1 )
break;
v16 = v15 - 1;
v11 = (uint32_t *)((char *)v6 + line);
v12 = (uint32_t *)((char *)v7 + ext);
v13 = wid_64;
do
{
v14 = *v11;
*v12 = v11[1];
v12[1] = v14;
v11 += 2;
v12 += 2;
--v13;
}
while ( v13 );
v6 = (uint32_t *)((char *)v11 + line);
v7 = (uint32_t *)((char *)v12 + ext);
v8 = v16 - 1;
}
while ( v16 != 1 );
}
//**************************************************************** //****************************************************************
// Size: 2, Format: 0 // Size: 2, Format: 0
@ -206,7 +172,7 @@ wxUint32 Load16bRGBA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int
if (height < 1) height = 1; if (height < 1) height = 1;
int ext = (real_width - (wid_64 << 2)) << 1; int ext = (real_width - (wid_64 << 2)) << 1;
asmLoad16bRGBA(src, dst, wid_64, height, line, ext); load16bRGBA((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext);
return (1 << 16) | GR_TEXFMT_ARGB_1555; return (1 << 16) | GR_TEXFMT_ARGB_1555;
} }
@ -221,7 +187,7 @@ wxUint32 Load16bIA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int li
if (height < 1) height = 1; if (height < 1) height = 1;
int ext = (real_width - (wid_64 << 2)) << 1; int ext = (real_width - (wid_64 << 2)) << 1;
asmLoad16bIA(src, dst, wid_64, height, line, ext); load16bIA((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext);
return (1 << 16) | GR_TEXFMT_ALPHA_INTENSITY_88; return (1 << 16) | GR_TEXFMT_ALPHA_INTENSITY_88;
} }

2440
Source/Glide64/TexLoad4b.h
View File

File diff suppressed because it is too large Load Diff

951
Source/Glide64/TexLoad8b.h
View File

@ -37,629 +37,336 @@
// //
//**************************************************************** //****************************************************************
extern "C" void __declspec(naked) asmLoad8bCI (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal) static inline void load8bCI(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext, uint16_t *pal)
{ {
_asm { uint8_t *v7;
push ebp uint32_t *v8;
mov ebp, esp int v9;
push ebx int v10;
push esi int v11;
push edi uint32_t v12;
uint32_t *v13;
uint32_t v14;
uint32_t *v15;
uint32_t v16;
uint32_t *v17;
uint32_t *v18;
int v19;
int v20;
uint32_t v21;
uint32_t v22;
uint32_t *v23;
uint32_t v24;
int v25;
int v26;
mov ebx,[pal] v7 = src;
mov esi,[src] v8 = (uint32_t *)dst;
mov edi,[dst] v9 = height;
mov ecx,[height] do
y_loop:
push ecx
mov ecx,[wid_64]
x_loop:
push ecx
mov eax,[esi] // read all 4 pixels
bswap eax
add esi,4
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,1
mov [edi],ecx
add edi,4
// }
// * copy
mov eax,[esi] // read all 4 pixels
bswap eax
add esi,4
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,1
mov [edi],ecx
add edi,4
// }
// *
pop ecx
dec ecx
jnz x_loop
pop ecx
dec ecx
jz near end_y_loop
push ecx
mov eax,esi
add eax,[line]
mov esi,[src]
sub eax,esi
and eax,0x7FF
add esi,eax
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
push ecx
mov eax,[esi+4] // read all 4 pixels
bswap eax
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,1
mov [edi],ecx
add edi,4
// }
// * copy
mov eax,[esi] // read all 4 pixels
bswap eax
mov edx,esi
add edx,8
mov esi,[src]
sub edx,esi
and edx,0x7FF
add esi,edx
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,1
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,1
mov [edi],ecx
add edi,4
// }
// *
pop ecx
dec ecx
jnz x_loop_2
mov eax,esi
add eax,[line]
mov esi,[src]
sub eax,esi
and eax,0x7FF
add esi,eax
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
}
}
extern "C" void __declspec(naked) asmLoad8bIA8 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext, wxUIntPtr pal)
{ {
_asm { v25 = v9;
push ebp v10 = wid_64;
mov ebp, esp do
push ebx
push esi
push edi
mov ebx,[pal]
mov esi,[src]
mov edi,[dst]
mov ecx,[height]
y_loop:
push ecx
mov ecx,[wid_64]
x_loop:
push ecx
mov eax,[esi] // read all 4 pixels
bswap eax
add esi,4
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,8
mov [edi],ecx
add edi,4
// }
// * copy
mov eax,[esi] // read all 4 pixels
bswap eax
add esi,4
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,8
mov [edi],ecx
add edi,4
// }
// *
pop ecx
dec ecx
jnz x_loop
pop ecx
dec ecx
jz near end_y_loop
push ecx
add esi,[line]
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
push ecx
mov eax,[esi+4] // read all 4 pixels
bswap eax
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,8
mov [edi],ecx
add edi,4
// }
// * copy
mov eax,[esi] // read all 4 pixels
bswap eax
add esi,8
mov edx,eax
// 1st dword output {
shr eax,15
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
mov eax,edx
shr eax,23
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
mov [edi],ecx
add edi,4
// }
// 2nd dword output {
mov eax,edx
shl eax,1
and eax,0x1FE
mov cx,[ebx+eax]
ror cx,8
shl ecx,16
shr edx,7
and edx,0x1FE
mov cx,[ebx+edx]
ror cx,8
mov [edi],ecx
add edi,4
// }
// *
pop ecx
dec ecx
jnz x_loop_2
add esi,[line]
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
}
}
extern "C" void __declspec(naked) asmLoad8bIA4 (wxUIntPtr src, wxUIntPtr dst, int wid_64, int height, int line, int ext)
{ {
_asm { v11 = v10;
push ebp v12 = bswap32(*(uint32_t *)v7);
mov ebp, esp v13 = (uint32_t *)(v7 + 4);
push ebx ALOWORD(v10) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 15) & 0x1FE)), 1);
push esi v14 = v10 << 16;
push edi ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 23) & 0x1FE)), 1);
*v8 = v14;
mov esi,[src] v15 = v8 + 1;
mov edi,[dst] ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v12 & 0x1FE)), 1);
mov ecx,[height] v14 <<= 16;
y_loop: ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 7) & 0x1FE)), 1);
push ecx *v15 = v14;
mov ecx,[wid_64] ++v15;
x_loop: v16 = bswap32(*v13);
mov eax,[esi] // read all 4 pixels v7 = (uint8_t *)(v13 + 1);
mov edx,eax ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 15) & 0x1FE)), 1);
v14 <<= 16;
shr eax,4 //all alpha ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 23) & 0x1FE)), 1);
shl edx,4 *v15 = v14;
and eax,0x0F0F0F0F ++v15;
and edx,0xF0F0F0F0 ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v16 & 0x1FE)), 1);
add esi,4 v14 <<= 16;
or eax,edx ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 7) & 0x1FE)), 1);
*v15 = v14;
mov [edi],eax // save dword v8 = v15 + 1;
add edi,4 v10 = v11 - 1;
mov eax,[esi] // read all 4 pixels
mov edx,eax
shr eax,4 //all alpha
shl edx,4
and eax,0x0F0F0F0F
and edx,0xF0F0F0F0
add esi,4
or eax,edx
mov [edi],eax // save dword
add edi,4
// *
dec ecx
jnz x_loop
pop ecx
dec ecx
jz end_y_loop
push ecx
add esi,[line]
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
mov eax,[esi+4] // read both pixels
mov edx,eax
shr eax,4 //all alpha
shl edx,4
and eax,0x0F0F0F0F
and edx,0xF0F0F0F0
or eax,edx
mov [edi],eax //save dword
add edi,4
mov eax,[esi] // read both pixels
add esi,8
mov edx,eax
shr eax,4 //all alpha
shl edx,4
and eax,0x0F0F0F0F
and edx,0xF0F0F0F0
or eax,edx
mov [edi],eax //save dword
add edi,4
// *
dec ecx
jnz x_loop_2
add esi,[line]
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
} }
} while ( v11 != 1 );
if ( v25 == 1 )
extern "C" void __declspec(naked) asmLoad8bI (wxUIntPtr src, int dst, wxUIntPtr wid_64, int height, int line, int ext) break;
v26 = v25 - 1;
v17 = (uint32_t *)&src[(line + (uintptr_t)v7 - (uintptr_t)src) & 0x7FF];
v18 = (uint32_t *)((char *)v8 + ext);
v19 = wid_64;
do
{ {
_asm { v20 = v19;
push ebp v21 = bswap32(v17[1]);
mov ebp, esp ALOWORD(v19) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 15) & 0x1FE)), 1);
push ebx v22 = v19 << 16;
push esi ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 23) & 0x1FE)), 1);
push edi *v18 = v22;
v23 = v18 + 1;
mov esi,[src] ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v21 & 0x1FE)), 1);
mov edi,[dst] v22 <<= 16;
mov ecx,[height] ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 7) & 0x1FE)), 1);
y_loop: *v23 = v22;
push ecx ++v23;
mov ecx,[wid_64] v24 = bswap32(*v17);
x_loop: v17 = (uint32_t *)&src[((uintptr_t)v17 + 8 - (uintptr_t)src) & 0x7FF];
mov eax,[esi] // read all 4 pixels ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 15) & 0x1FE)), 1);
add esi,4 v22 <<= 16;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 23) & 0x1FE)), 1);
mov [edi],eax // save dword *v23 = v22;
add edi,4 ++v23;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v24 & 0x1FE)), 1);
mov eax,[esi] // read all 4 pixels v22 <<= 16;
add esi,4 ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 7) & 0x1FE)), 1);
*v23 = v22;
mov [edi],eax // save dword v18 = v23 + 1;
add edi,4 v19 = v20 - 1;
// *
dec ecx
jnz x_loop
pop ecx
dec ecx
jz end_y_loop
push ecx
add esi,[line]
add edi,[ext]
mov ecx,[wid_64]
x_loop_2:
mov eax,[esi+4] // read both pixels
mov [edi],eax //save dword
add edi,4
mov eax,[esi] // read both pixels
add esi,8
mov [edi],eax //save dword
add edi,4
// *
dec ecx
jnz x_loop_2
add esi,[line]
add edi,[ext]
pop ecx
dec ecx
jnz y_loop
end_y_loop:
pop edi
pop esi
pop ebx
mov esp, ebp
pop ebp
ret
} }
while ( v20 != 1 );
v7 = &src[(line + (uintptr_t)v17 - (uintptr_t)src) & 0x7FF];
v8 = (uint32_t *)((char *)v18 + ext);
v9 = v26 - 1;
}
while ( v26 != 1 );
}
static inline void load8bIA8(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext, uint16_t *pal)
{
uint32_t *v7;
uint32_t *v8;
int v9;
int v10;
int v11;
uint32_t v12;
uint32_t *v13;
uint32_t v14;
uint32_t *v15;
uint32_t v16;
uint32_t *v17;
uint32_t *v18;
int v19;
int v20;
uint32_t v21;
uint32_t v22;
uint32_t *v23;
uint32_t v24;
int v25;
int v26;
v7 = (uint32_t *)src;
v8 = (uint32_t *)dst;
v9 = height;
do
{
v25 = v9;
v10 = wid_64;
do
{
v11 = v10;
v12 = bswap32(*v7);
v13 = v7 + 1;
ALOWORD(v10) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 15) & 0x1FE)), 8);
v14 = v10 << 16;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 23) & 0x1FE)), 8);
*v8 = v14;
v15 = v8 + 1;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v12 & 0x1FE)), 8);
v14 <<= 16;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v12 >> 7) & 0x1FE)), 8);
*v15 = v14;
++v15;
v16 = bswap32(*v13);
v7 = v13 + 1;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 15) & 0x1FE)), 8);
v14 <<= 16;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 23) & 0x1FE)), 8);
*v15 = v14;
++v15;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v16 & 0x1FE)), 8);
v14 <<= 16;
ALOWORD(v14) = __ROR__(*(uint16_t *)((char *)pal + ((v16 >> 7) & 0x1FE)), 8);
*v15 = v14;
v8 = v15 + 1;
v10 = v11 - 1;
}
while ( v11 != 1 );
if ( v25 == 1 )
break;
v26 = v25 - 1;
v17 = (uint32_t *)((char *)v7 + line);
v18 = (uint32_t *)((char *)v8 + ext);
v19 = wid_64;
do
{
v20 = v19;
v21 = bswap32(v17[1]);
ALOWORD(v19) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 15) & 0x1FE)), 8);
v22 = v19 << 16;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 23) & 0x1FE)), 8);
*v18 = v22;
v23 = v18 + 1;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v21 & 0x1FE)), 8);
v22 <<= 16;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v21 >> 7) & 0x1FE)), 8);
*v23 = v22;
++v23;
v24 = bswap32(*v17);
v17 += 2;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 15) & 0x1FE)), 8);
v22 <<= 16;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 23) & 0x1FE)), 8);
*v23 = v22;
++v23;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + (2 * (uint16_t)v24 & 0x1FE)), 8);
v22 <<= 16;
ALOWORD(v22) = __ROR__(*(uint16_t *)((char *)pal + ((v24 >> 7) & 0x1FE)), 8);
*v23 = v22;
v18 = v23 + 1;
v19 = v20 - 1;
}
while ( v20 != 1 );
v7 = (uint32_t *)((char *)v17 + line);
v8 = (uint32_t *)((char *)v18 + ext);
v9 = v26 - 1;
}
while ( v26 != 1 );
}
static inline void load8bIA4(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext)
{
uint32_t *v6;
uint32_t *v7;
int v8;
int v9;
uint32_t v10;
uint32_t v11;
uint32_t *v12;
uint32_t *v13;
uint32_t v14;
uint32_t v15;
uint32_t *v16;
uint32_t *v17;
int v18;
uint32_t *v19;
uint32_t v20;
int v21;
int v22;
v6 = (uint32_t *)src;
v7 = (uint32_t *)dst;
v8 = height;
do
{
v21 = v8;
v9 = wid_64;
do
{
v10 = *v6;
v11 = (*v6 >> 4) & 0xF0F0F0F;
v12 = v6 + 1;
*v7 = (16 * v10 & 0xF0F0F0F0) | v11;
v13 = v7 + 1;
v14 = (*v12 >> 4) & 0xF0F0F0F;
v15 = 16 * *v12 & 0xF0F0F0F0;
v6 = v12 + 1;
*v13 = v15 | v14;
v7 = v13 + 1;
--v9;
}
while ( v9 );
if ( v21 == 1 )
break;
v22 = v21 - 1;
v16 = (uint32_t *)((char *)v6 + line);
v17 = (uint32_t *)((char *)v7 + ext);
v18 = wid_64;
do
{
*v17 = (16 * v16[1] & 0xF0F0F0F0) | ((v16[1] >> 4) & 0xF0F0F0F);
v19 = v17 + 1;
v20 = *v16;
v16 += 2;
*v19 = (16 * v20 & 0xF0F0F0F0) | ((v20 >> 4) & 0xF0F0F0F);
v17 = v19 + 1;
--v18;
}
while ( v18 );
v6 = (uint32_t *)((char *)v16 + line);
v7 = (uint32_t *)((char *)v17 + ext);
v8 = v22 - 1;
}
while ( v22 != 1 );
}
static inline void load8bI(uint8_t *src, uint8_t *dst, int wid_64, int height, int line, int ext)
{
uint32_t *v6;
uint32_t *v7;
int v8;
int v9;
uint32_t v10;
uint32_t *v11;
uint32_t *v12;
uint32_t v13;
uint32_t *v14;
uint32_t *v15;
int v16;
uint32_t *v17;
uint32_t v18;
int v19;
int v20;
v6 = (uint32_t *)src;
v7 = (uint32_t *)dst;
v8 = height;
do
{
v19 = v8;
v9 = wid_64;
do
{
v10 = *v6;
v11 = v6 + 1;
*v7 = v10;
v12 = v7 + 1;
v13 = *v11;
v6 = v11 + 1;
*v12 = v13;
v7 = v12 + 1;
--v9;
}
while ( v9 );
if ( v19 == 1 )
break;
v20 = v19 - 1;
v14 = (uint32_t *)((char *)v6 + line);
v15 = (uint32_t *)((char *)v7 + ext);
v16 = wid_64;
do
{
*v15 = v14[1];
v17 = v15 + 1;
v18 = *v14;
v14 += 2;
*v17 = v18;
v15 = v17 + 1;
--v16;
}
while ( v16 );
v6 = (uint32_t *)((char *)v14 + line);
v7 = (uint32_t *)((char *)v15 + ext);
v8 = v20 - 1;
}
while ( v20 != 1 );
} }
//**************************************************************** //****************************************************************
@ -671,21 +378,21 @@ wxUint32 Load8bCI (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int lin
if (wid_64 < 1) wid_64 = 1; if (wid_64 < 1) wid_64 = 1;
if (height < 1) height = 1; if (height < 1) height = 1;
int ext = (real_width - (wid_64 << 3)); int ext = (real_width - (wid_64 << 3));
wxUIntPtr pal = wxPtrToUInt(rdp.pal_8); unsigned short * pal = rdp.pal_8;
switch (rdp.tlut_mode) { switch (rdp.tlut_mode) {
case 0: //palette is not used case 0: //palette is not used
//in tlut DISABLE mode load CI texture as plain intensity texture instead of palette dereference. //in tlut DISABLE mode load CI texture as plain intensity texture instead of palette dereference.
//Thanks to angrylion for the advice //Thanks to angrylion for the advice
asmLoad8bI (src, dst, wid_64, height, line, ext); load8bI ((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext);
return /*(0 << 16) | */GR_TEXFMT_ALPHA_8; return /*(0 << 16) | */GR_TEXFMT_ALPHA_8;
case 2: //color palette case 2: //color palette
ext <<= 1; ext <<= 1;
asmLoad8bCI (src, dst, wid_64, height, line, ext, pal); load8bCI ((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext, pal);
return (1 << 16) | GR_TEXFMT_ARGB_1555; return (1 << 16) | GR_TEXFMT_ARGB_1555;
default: //IA palette default: //IA palette
ext <<= 1; ext <<= 1;
asmLoad8bIA8 (src, dst, wid_64, height, line, ext, pal); load8bIA8 ((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext, pal);
return (1 << 16) | GR_TEXFMT_ALPHA_INTENSITY_88; return (1 << 16) | GR_TEXFMT_ALPHA_INTENSITY_88;
} }
} }
@ -702,7 +409,7 @@ wxUint32 Load8bIA (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int lin
if (wid_64 < 1) wid_64 = 1; if (wid_64 < 1) wid_64 = 1;
if (height < 1) height = 1; if (height < 1) height = 1;
int ext = (real_width - (wid_64 << 3)); int ext = (real_width - (wid_64 << 3));
asmLoad8bIA4 (src, dst, wid_64, height, line, ext); load8bIA4 ((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext);
return /*(0 << 16) | */GR_TEXFMT_ALPHA_INTENSITY_44; return /*(0 << 16) | */GR_TEXFMT_ALPHA_INTENSITY_44;
} }
@ -718,6 +425,6 @@ wxUint32 Load8bI (wxUIntPtr dst, wxUIntPtr src, int wid_64, int height, int line
if (wid_64 < 1) wid_64 = 1; if (wid_64 < 1) wid_64 = 1;
if (height < 1) height = 1; if (height < 1) height = 1;
int ext = (real_width - (wid_64 << 3)); int ext = (real_width - (wid_64 << 3));
asmLoad8bI (src, dst, wid_64, height, line, ext); load8bI ((uint8_t *)src, (uint8_t *)dst, wid_64, height, line, ext);
return /*(0 << 16) | */GR_TEXFMT_ALPHA_8; return /*(0 << 16) | */GR_TEXFMT_ALPHA_8;
} }