Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Speed optimizations for 2xBR, xBR-lv1 and DDT. #1072

Merged
merged 1 commit into from
Sep 23, 2023
Merged

Speed optimizations for 2xBR, xBR-lv1 and DDT. #1072

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
66 changes: 31 additions & 35 deletions source/filter.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -667,14 +667,13 @@ static const uint16 lb_mask = LB_MASK565;


#define XBR(PE, PI, PH, PF, PG, PC, PD, PB, PA, N0, N1, N2, N3) \
irlv1 = (PE!=PH && PE!=PF); \
if ( irlv1 )\
if ( PE!=PH && PE!=PF )\
{\
wd1 = (df(PE,PC)+df(PE,PG))+(df(PH,PF)<<2); \
wd2 = (df(PH,PD)+df(PF,PB))+(df(PE,PI)<<2); \
if (wd1<wd2)\
wd1 = df(PH,PF); \
wd2 = df(PE,PI); \
if ((wd1<<1)<wd2)\
{\
if ( ( !eq(PF,PB) && !eq(PF,PC) || !eq(PH,PD) && !eq(PH,PG) || eq(PE,PG) || eq(PE,PC)) )\
if ( !eq(PF,PB) && !eq(PF,PC) || !eq(PH,PD) && !eq(PH,PG) || eq(PE,PG) || eq(PE,PC) )\
{\
dFG=df(PF,PG); dHC=df(PH,PC); \
irlv2u = (PE!=PC && PB!=PC); irlv2l = (PE!=PG && PD!=PG); px = (df(PE,PF) <= df(PE,PH)) ? PF : PH; \
Expand Down Expand Up @@ -703,12 +702,13 @@ static const uint16 lb_mask = LB_MASK565;
}\
}\


#define XBRLV1(PE, PI, PH, PF, PG, PC, PD, PB, PA, N0, N1, N2, N3) \
irlv1 = (PE!=PH && PE!=PF); \
if ( irlv1 )\
{\
wd1 = ((df(PE,PC)+df(PE,PG))+(df(PH,PF)<<1)); \
wd2 = ( df(PH,PD)+df(PF,PB))+(df(PE,PI)<<1); \
wd1 = df(PH,PF); \
wd2 = df(PE,PI); \
if (((wd1<<1)<wd2) && eq(PB,PD) && PB!=PF && PD!=PH)\
{\
px = (df(PE,PF) <= df(PE,PH)) ? PF : PH; \
Expand All @@ -722,7 +722,7 @@ static const uint16 lb_mask = LB_MASK565;
}\


#define DDT(PE, PI, PH, PF, PG, PC, PD, PB, PA, N0, N1, N2, N3) \
#define DDT(PE, PI, PH, PF, N0, N1, N2, N3) \
wd1 = (df(PH,PF)); \
wd2 = (df(PE,PI)); \
if (wd1>wd2)\
Expand Down Expand Up @@ -803,10 +803,13 @@ void Render2xBR (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPitch,

Ep[E0] = Ep[E1] = Ep[E2] = Ep[E3] = E; // 0, 1, 2, 3

XBR( E, I, H, F, G, C, D, B, A, E0, E1, E2, E3);
XBR( E, C, F, B, I, A, H, D, G, E2, E0, E3, E1);
XBR( E, A, B, D, C, G, F, H, I, E3, E2, E1, E0);
XBR( E, G, D, H, A, I, B, F, C, E1, E3, E0, E2);
if ( (E!=F || E!=D) && (E!=H || E!=B) )
{
XBR( E, I, H, F, G, C, D, B, A, E0, E1, E2, E3);
XBR( E, C, F, B, I, A, H, D, G, E2, E0, E3, E1);
XBR( E, A, B, D, C, G, F, H, I, E3, E2, E1, E0);
XBR( E, G, D, H, A, I, B, F, C, E1, E3, E0, E2);
}

A= B; B=C;
D= E; E=F;
Expand All @@ -828,8 +831,7 @@ void Render2xBRlv1 (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPit
}

uint32 wd1, wd2;
uint32 irlv1, irlv2u, irlv2l;
uint32 dFG, dHC;
uint32 irlv1;
uint32 E0, E1, E2, E3;

uint16 A, B, C, D, E, F, G, H, I, px;
Expand Down Expand Up @@ -863,11 +865,14 @@ void Render2xBRlv1 (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPit
E3 = (i << 1) + nextlineDst + 1;

Ep[E0] = Ep[E1] = Ep[E2] = Ep[E3] = E; // 0, 1, 2, 3

XBRLV1( E, I, H, F, G, C, D, B, A, E0, E1, E2, E3);
XBRLV1( E, C, F, B, I, A, H, D, G, E2, E0, E3, E1);
XBRLV1( E, A, B, D, C, G, F, H, I, E3, E2, E1, E0);
XBRLV1( E, G, D, H, A, I, B, F, C, E1, E3, E0, E2);

if ( (E!=F || E!=D) && (E!=H || E!=B) )
{
XBRLV1( E, I, H, F, G, C, D, B, A, E0, E1, E2, E3);
XBRLV1( E, C, F, B, I, A, H, D, G, E2, E0, E3, E1);
XBRLV1( E, A, B, D, C, G, F, H, I, E3, E2, E1, E0);
XBRLV1( E, G, D, H, A, I, B, F, C, E1, E3, E0, E2);
}

A= B; B=C;
D= E; E=F;
Expand All @@ -879,6 +884,7 @@ void Render2xBRlv1 (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPit
}
}


template<int GuiScale>
void RenderDDT (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPitch, int width, int height)
{
Expand All @@ -889,11 +895,9 @@ void RenderDDT (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPitch,
}

uint32 wd1, wd2;
uint32 irlv1, irlv2u, irlv2l;
uint32 dFG, dHC;
uint32 E0, E1, E2, E3;

uint16 A, B, C, D, E, F, G, H, I, px, aux;
uint16 E, F, H, I, aux;

uint32 nextlineSrc = srcPitch / sizeof(uint16);
uint16 *p = (uint16 *)srcPtr;
Expand All @@ -903,18 +907,10 @@ void RenderDDT (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPitch,

while (height--) {

A = *(p - 1 - nextlineSrc);
B = *(p - nextlineSrc);

D = *(p - 1);
E = *(p);

G = *(p - 1 + nextlineSrc);
H = *(p + nextlineSrc);


for (int i = 0; i < width; i++) {
C = *(p + i + 1 - nextlineSrc);
F = *(p + i + 1);
I = *(p + i + 1 + nextlineSrc);

Expand All @@ -927,16 +923,16 @@ void RenderDDT (uint8 *srcPtr, uint32 srcPitch, uint8 *dstPtr, uint32 dstPitch,

if (E!=F || E!=H || F!=I || H!=I)
{
DDT( E, I, H, F, G, C, D, B, A, E0, E1, E2, E3);
DDT( E, I, H, F, E0, E1, E2, E3);
}

A= B; B=C;
D= E; E=F;
G= H; H=I;
E=F;
H=I;
}

p += nextlineSrc;
Ep += nextlineDst << 1;
}
}