Sprinter/mame
2
0
Fork 0
mirror of https://github.com/holub/mame synced 2025-06-06 12:53:46 +03:00
Code Issues Actions Packages Projects Releases Wiki Activity

naomi: support texture stride selection [O. Galibert]

Browse Source
This commit is contained in:
Olivier Galibert 2011-01-10 19:04:00 +00:00
parent 0986367da7
commit 64802b1787
1 changed files with 12 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
src/mame/video/dc.c
View File

@ -152,7 +152,7 @@ static void render_to_accumulation_buffer(running_machine *machine,bitmap_t *bit
typedef struct texinfo { typedef struct texinfo {
UINT32 address, vqbase; UINT32 address, vqbase;
int textured, sizex, sizey, sizes, pf, palette, mode, mipmapped, blend_mode, filter_mode, flip_u, flip_v; int textured, sizex, sizey, stride, sizes, pf, palette, mode, mipmapped, blend_mode, filter_mode, flip_u, flip_v;
UINT32 (*r)(struct texinfo *t, float x, float y); UINT32 (*r)(struct texinfo *t, float x, float y);
UINT32 (*blend)(UINT32 s, UINT32 d); UINT32 (*blend)(UINT32 s, UINT32 d);
@ -435,7 +435,7 @@ INLINE UINT32 tex_r_yuv_n(texinfo *t, float x, float y)
{ {
int xt = ((int)x) & (t->sizex-1); int xt = ((int)x) & (t->sizex-1);
int yt = ((int)y) & (t->sizey-1); int yt = ((int)y) & (t->sizey-1);
int addrp = t->address + (t->sizex*yt + (xt & ~1))*2; int addrp = t->address + (t->stride*yt + (xt & ~1))*2;
UINT16 c1 = *(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp)); UINT16 c1 = *(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp));
UINT16 c2 = *(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp+2)); UINT16 c2 = *(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp+2));
return cv_yuv(c1, c2, xt); return cv_yuv(c1, c2, xt);
@ -445,7 +445,7 @@ INLINE UINT32 tex_r_1555_n(texinfo *t, float x, float y)
{ {
int xt = ((int)x) & (t->sizex-1); int xt = ((int)x) & (t->sizex-1);
int yt = ((int)y) & (t->sizey-1); int yt = ((int)y) & (t->sizey-1);
int addrp = t->address + (t->sizex*yt + xt)*2; int addrp = t->address + (t->stride*yt + xt)*2;
return cv_1555z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp))); return cv_1555z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp)));
} }
@ -470,7 +470,7 @@ INLINE UINT32 tex_r_565_n(texinfo *t, float x, float y)
{ {
int xt = ((int)x) & (t->sizex-1); int xt = ((int)x) & (t->sizex-1);
int yt = ((int)y) & (t->sizey-1); int yt = ((int)y) & (t->sizey-1);
int addrp = t->address + (t->sizex*yt + xt)*2; int addrp = t->address + (t->stride*yt + xt)*2;
return cv_565z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp))); return cv_565z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp)));
} }
@ -495,7 +495,7 @@ INLINE UINT32 tex_r_4444_n(texinfo *t, float x, float y)
{ {
int xt = ((int)x) & (t->sizex-1); int xt = ((int)x) & (t->sizex-1);
int yt = ((int)y) & (t->sizey-1); int yt = ((int)y) & (t->sizey-1);
int addrp = t->address + (t->sizex*yt + xt)*2; int addrp = t->address + (t->stride*yt + xt)*2;
return cv_4444z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp))); return cv_4444z(*(UINT16 *)(((UINT8 *)dc_texture_ram) + WORD_XOR_LE(addrp)));
} }
@ -691,7 +691,6 @@ static void tex_get_info(texinfo *t, pvrta_state *sa)
t->pf = sa->pixelformat; t->pf = sa->pixelformat;
t->palette = 0; t->palette = 0;
t->mode = (sa->vqcompressed<<1); t->mode = (sa->vqcompressed<<1);
// scanorder is ignored for palettized textures (palettized textures are ALWAYS twiddled) // scanorder is ignored for palettized textures (palettized textures are ALWAYS twiddled)
@ -705,24 +704,8 @@ static void tex_get_info(texinfo *t, pvrta_state *sa)
t->mode |= sa->scanorder; t->mode |= sa->scanorder;
} }
/* When scan order is 0 stride select is ignored */ /* When scan order is 1 (non-twiddled) mipmap is ignored */
/* When scan order is 1 mipmap is ignored */ t->mipmapped = t->mode & 1 ? 0 : sa->mipmapped;
if (t->mode&1)
{
/* scan order is 1 (non-twiddled tezture), use stride select if specified*/
t->mode |= (sa->strideselect<<2);
/* scan order is 1 (non-twiddled tezture), ignore mipmaps */
t->mipmapped = 0;
}
else
{
/* scan order is 0 (twiddled tezture), ignore stride select*/
//t->mode += (sa->strideselect<<2);
/* scan order is 0 (twiddled tezture), use mipmap if specified */
t->mipmapped = sa->mipmapped;
}
// Mipmapped textures are always square, ignore v size // Mipmapped textures are always square, ignore v size
if (t->mipmapped) if (t->mipmapped)
@ -737,7 +720,10 @@ static void tex_get_info(texinfo *t, pvrta_state *sa)
t->sizex = 1 << (3+((t->sizes >> 3) & 7)); t->sizex = 1 << (3+((t->sizes >> 3) & 7));
t->sizey = 1 << (3+(t->sizes & 7)); t->sizey = 1 << (3+(t->sizes & 7));
/* Stride select is used only in the non-twiddled case */
t->stride = (t->mode & 1) && sa->strideselect ? t->sizex : (pvrta_regs[TEXT_CONTROL] & 0x1f) << 5;
t->blend_mode = sa->blend_mode; t->blend_mode = sa->blend_mode;
t->filter_mode = sa->filtermode; t->filter_mode = sa->filtermode;
@ -750,7 +736,7 @@ static void tex_get_info(texinfo *t, pvrta_state *sa)
t->vqbase = t->address; t->vqbase = t->address;
t->blend = sa->use_alpha ? blend_functions[t->blend_mode] : bl10; t->blend = sa->use_alpha ? blend_functions[t->blend_mode] : bl10;
// fprintf(stderr, "tex %d %d %d %d\n", t->pf, t->mode, pvrta_regs[PAL_RAM_CTRL], t->mipmapped); // fprintf(stderr, "tex %d %d %d %d\n", t->pf, t->mode, pvrta_regs[PAL_RAM_CTRL], t->mipmapped);
switch(t->pf) { switch(t->pf) {
case 0: // 1555 case 0: // 1555