+#define SRC(x,y) src[(x)+(y)*stride]
+#define PL(y) \
+ const int l##y = (SRC(-1,y-1) + 2*SRC(-1,y) + SRC(-1,y+1) + 2) >> 2;
+#define PREDICT_8x8_LOAD_LEFT \
+ const int l0 = ((has_topleft ? SRC(-1,-1) : SRC(-1,0)) \
+ + 2*SRC(-1,0) + SRC(-1,1) + 2) >> 2; \
+ PL(1) PL(2) PL(3) PL(4) PL(5) PL(6) \
+ const int l7 attribute_unused = (SRC(-1,6) + 3*SRC(-1,7) + 2) >> 2
+
+#define PT(x) \
+ const int t##x = (SRC(x-1,-1) + 2*SRC(x,-1) + SRC(x+1,-1) + 2) >> 2;
+#define PREDICT_8x8_LOAD_TOP \
+ const int t0 = ((has_topleft ? SRC(-1,-1) : SRC(0,-1)) \
+ + 2*SRC(0,-1) + SRC(1,-1) + 2) >> 2; \
+ PT(1) PT(2) PT(3) PT(4) PT(5) PT(6) \
+ const int t7 attribute_unused = ((has_topright ? SRC(8,-1) : SRC(7,-1)) \
+ + 2*SRC(7,-1) + SRC(6,-1) + 2) >> 2
+
+#define PTR(x) \
+ t##x = (SRC(x-1,-1) + 2*SRC(x,-1) + SRC(x+1,-1) + 2) >> 2;
+#define PREDICT_8x8_LOAD_TOPRIGHT \
+ int t8, t9, t10, t11, t12, t13, t14, t15; \
+ if(has_topright) { \
+ PTR(8) PTR(9) PTR(10) PTR(11) PTR(12) PTR(13) PTR(14) \
+ t15 = (SRC(14,-1) + 3*SRC(15,-1) + 2) >> 2; \
+ } else t8=t9=t10=t11=t12=t13=t14=t15= SRC(7,-1);
+
+#define PREDICT_8x8_LOAD_TOPLEFT \
+ const int lt = (SRC(-1,0) + 2*SRC(-1,-1) + SRC(0,-1) + 2) >> 2
+
+#define PREDICT_8x8_DC(v) \
+ int y; \
+ for( y = 0; y < 8; y++ ) { \
+ ((uint32_t*)src)[0] = \
+ ((uint32_t*)src)[1] = v; \
+ src += stride; \
+ }
+
+static void pred8x8l_128_dc_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_DC(0x80808080);
+}
+static void pred8x8l_left_dc_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_LEFT;
+ const uint32_t dc = ((l0+l1+l2+l3+l4+l5+l6+l7+4) >> 3) * 0x01010101;
+ PREDICT_8x8_DC(dc);
+}
+static void pred8x8l_top_dc_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ const uint32_t dc = ((t0+t1+t2+t3+t4+t5+t6+t7+4) >> 3) * 0x01010101;
+ PREDICT_8x8_DC(dc);
+}
+static void pred8x8l_dc_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_LEFT;
+ PREDICT_8x8_LOAD_TOP;
+ const uint32_t dc = ((l0+l1+l2+l3+l4+l5+l6+l7
+ +t0+t1+t2+t3+t4+t5+t6+t7+8) >> 4) * 0x01010101;
+ PREDICT_8x8_DC(dc);
+}
+static void pred8x8l_horizontal_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_LEFT;
+#define ROW(y) ((uint32_t*)(src+y*stride))[0] =\
+ ((uint32_t*)(src+y*stride))[1] = 0x01010101 * l##y
+ ROW(0); ROW(1); ROW(2); ROW(3); ROW(4); ROW(5); ROW(6); ROW(7);
+#undef ROW
+}
+static void pred8x8l_vertical_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ int y;
+ PREDICT_8x8_LOAD_TOP;
+ src[0] = t0;
+ src[1] = t1;
+ src[2] = t2;
+ src[3] = t3;
+ src[4] = t4;
+ src[5] = t5;
+ src[6] = t6;
+ src[7] = t7;
+ for( y = 1; y < 8; y++ )
+ *(uint64_t*)(src+y*stride) = *(uint64_t*)src;
+}
+static void pred8x8l_down_left_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ PREDICT_8x8_LOAD_TOPRIGHT;
+ SRC(0,0)= (t0 + 2*t1 + t2 + 2) >> 2;
+ SRC(0,1)=SRC(1,0)= (t1 + 2*t2 + t3 + 2) >> 2;
+ SRC(0,2)=SRC(1,1)=SRC(2,0)= (t2 + 2*t3 + t4 + 2) >> 2;
+ SRC(0,3)=SRC(1,2)=SRC(2,1)=SRC(3,0)= (t3 + 2*t4 + t5 + 2) >> 2;
+ SRC(0,4)=SRC(1,3)=SRC(2,2)=SRC(3,1)=SRC(4,0)= (t4 + 2*t5 + t6 + 2) >> 2;
+ SRC(0,5)=SRC(1,4)=SRC(2,3)=SRC(3,2)=SRC(4,1)=SRC(5,0)= (t5 + 2*t6 + t7 + 2) >> 2;
+ SRC(0,6)=SRC(1,5)=SRC(2,4)=SRC(3,3)=SRC(4,2)=SRC(5,1)=SRC(6,0)= (t6 + 2*t7 + t8 + 2) >> 2;
+ SRC(0,7)=SRC(1,6)=SRC(2,5)=SRC(3,4)=SRC(4,3)=SRC(5,2)=SRC(6,1)=SRC(7,0)= (t7 + 2*t8 + t9 + 2) >> 2;
+ SRC(1,7)=SRC(2,6)=SRC(3,5)=SRC(4,4)=SRC(5,3)=SRC(6,2)=SRC(7,1)= (t8 + 2*t9 + t10 + 2) >> 2;
+ SRC(2,7)=SRC(3,6)=SRC(4,5)=SRC(5,4)=SRC(6,3)=SRC(7,2)= (t9 + 2*t10 + t11 + 2) >> 2;
+ SRC(3,7)=SRC(4,6)=SRC(5,5)=SRC(6,4)=SRC(7,3)= (t10 + 2*t11 + t12 + 2) >> 2;
+ SRC(4,7)=SRC(5,6)=SRC(6,5)=SRC(7,4)= (t11 + 2*t12 + t13 + 2) >> 2;
+ SRC(5,7)=SRC(6,6)=SRC(7,5)= (t12 + 2*t13 + t14 + 2) >> 2;
+ SRC(6,7)=SRC(7,6)= (t13 + 2*t14 + t15 + 2) >> 2;
+ SRC(7,7)= (t14 + 3*t15 + 2) >> 2;
+}
+static void pred8x8l_down_right_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ PREDICT_8x8_LOAD_LEFT;
+ PREDICT_8x8_LOAD_TOPLEFT;
+ SRC(0,7)= (l7 + 2*l6 + l5 + 2) >> 2;
+ SRC(0,6)=SRC(1,7)= (l6 + 2*l5 + l4 + 2) >> 2;
+ SRC(0,5)=SRC(1,6)=SRC(2,7)= (l5 + 2*l4 + l3 + 2) >> 2;
+ SRC(0,4)=SRC(1,5)=SRC(2,6)=SRC(3,7)= (l4 + 2*l3 + l2 + 2) >> 2;
+ SRC(0,3)=SRC(1,4)=SRC(2,5)=SRC(3,6)=SRC(4,7)= (l3 + 2*l2 + l1 + 2) >> 2;
+ SRC(0,2)=SRC(1,3)=SRC(2,4)=SRC(3,5)=SRC(4,6)=SRC(5,7)= (l2 + 2*l1 + l0 + 2) >> 2;
+ SRC(0,1)=SRC(1,2)=SRC(2,3)=SRC(3,4)=SRC(4,5)=SRC(5,6)=SRC(6,7)= (l1 + 2*l0 + lt + 2) >> 2;
+ SRC(0,0)=SRC(1,1)=SRC(2,2)=SRC(3,3)=SRC(4,4)=SRC(5,5)=SRC(6,6)=SRC(7,7)= (l0 + 2*lt + t0 + 2) >> 2;
+ SRC(1,0)=SRC(2,1)=SRC(3,2)=SRC(4,3)=SRC(5,4)=SRC(6,5)=SRC(7,6)= (lt + 2*t0 + t1 + 2) >> 2;
+ SRC(2,0)=SRC(3,1)=SRC(4,2)=SRC(5,3)=SRC(6,4)=SRC(7,5)= (t0 + 2*t1 + t2 + 2) >> 2;
+ SRC(3,0)=SRC(4,1)=SRC(5,2)=SRC(6,3)=SRC(7,4)= (t1 + 2*t2 + t3 + 2) >> 2;
+ SRC(4,0)=SRC(5,1)=SRC(6,2)=SRC(7,3)= (t2 + 2*t3 + t4 + 2) >> 2;
+ SRC(5,0)=SRC(6,1)=SRC(7,2)= (t3 + 2*t4 + t5 + 2) >> 2;
+ SRC(6,0)=SRC(7,1)= (t4 + 2*t5 + t6 + 2) >> 2;
+ SRC(7,0)= (t5 + 2*t6 + t7 + 2) >> 2;
+
+}
+static void pred8x8l_vertical_right_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ PREDICT_8x8_LOAD_LEFT;
+ PREDICT_8x8_LOAD_TOPLEFT;
+ SRC(0,6)= (l5 + 2*l4 + l3 + 2) >> 2;
+ SRC(0,7)= (l6 + 2*l5 + l4 + 2) >> 2;
+ SRC(0,4)=SRC(1,6)= (l3 + 2*l2 + l1 + 2) >> 2;
+ SRC(0,5)=SRC(1,7)= (l4 + 2*l3 + l2 + 2) >> 2;
+ SRC(0,2)=SRC(1,4)=SRC(2,6)= (l1 + 2*l0 + lt + 2) >> 2;
+ SRC(0,3)=SRC(1,5)=SRC(2,7)= (l2 + 2*l1 + l0 + 2) >> 2;
+ SRC(0,1)=SRC(1,3)=SRC(2,5)=SRC(3,7)= (l0 + 2*lt + t0 + 2) >> 2;
+ SRC(0,0)=SRC(1,2)=SRC(2,4)=SRC(3,6)= (lt + t0 + 1) >> 1;
+ SRC(1,1)=SRC(2,3)=SRC(3,5)=SRC(4,7)= (lt + 2*t0 + t1 + 2) >> 2;
+ SRC(1,0)=SRC(2,2)=SRC(3,4)=SRC(4,6)= (t0 + t1 + 1) >> 1;
+ SRC(2,1)=SRC(3,3)=SRC(4,5)=SRC(5,7)= (t0 + 2*t1 + t2 + 2) >> 2;
+ SRC(2,0)=SRC(3,2)=SRC(4,4)=SRC(5,6)= (t1 + t2 + 1) >> 1;
+ SRC(3,1)=SRC(4,3)=SRC(5,5)=SRC(6,7)= (t1 + 2*t2 + t3 + 2) >> 2;
+ SRC(3,0)=SRC(4,2)=SRC(5,4)=SRC(6,6)= (t2 + t3 + 1) >> 1;
+ SRC(4,1)=SRC(5,3)=SRC(6,5)=SRC(7,7)= (t2 + 2*t3 + t4 + 2) >> 2;
+ SRC(4,0)=SRC(5,2)=SRC(6,4)=SRC(7,6)= (t3 + t4 + 1) >> 1;
+ SRC(5,1)=SRC(6,3)=SRC(7,5)= (t3 + 2*t4 + t5 + 2) >> 2;
+ SRC(5,0)=SRC(6,2)=SRC(7,4)= (t4 + t5 + 1) >> 1;
+ SRC(6,1)=SRC(7,3)= (t4 + 2*t5 + t6 + 2) >> 2;
+ SRC(6,0)=SRC(7,2)= (t5 + t6 + 1) >> 1;
+ SRC(7,1)= (t5 + 2*t6 + t7 + 2) >> 2;
+ SRC(7,0)= (t6 + t7 + 1) >> 1;
+}
+static void pred8x8l_horizontal_down_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ PREDICT_8x8_LOAD_LEFT;
+ PREDICT_8x8_LOAD_TOPLEFT;
+ SRC(0,7)= (l6 + l7 + 1) >> 1;
+ SRC(1,7)= (l5 + 2*l6 + l7 + 2) >> 2;
+ SRC(0,6)=SRC(2,7)= (l5 + l6 + 1) >> 1;
+ SRC(1,6)=SRC(3,7)= (l4 + 2*l5 + l6 + 2) >> 2;
+ SRC(0,5)=SRC(2,6)=SRC(4,7)= (l4 + l5 + 1) >> 1;
+ SRC(1,5)=SRC(3,6)=SRC(5,7)= (l3 + 2*l4 + l5 + 2) >> 2;
+ SRC(0,4)=SRC(2,5)=SRC(4,6)=SRC(6,7)= (l3 + l4 + 1) >> 1;
+ SRC(1,4)=SRC(3,5)=SRC(5,6)=SRC(7,7)= (l2 + 2*l3 + l4 + 2) >> 2;
+ SRC(0,3)=SRC(2,4)=SRC(4,5)=SRC(6,6)= (l2 + l3 + 1) >> 1;
+ SRC(1,3)=SRC(3,4)=SRC(5,5)=SRC(7,6)= (l1 + 2*l2 + l3 + 2) >> 2;
+ SRC(0,2)=SRC(2,3)=SRC(4,4)=SRC(6,5)= (l1 + l2 + 1) >> 1;
+ SRC(1,2)=SRC(3,3)=SRC(5,4)=SRC(7,5)= (l0 + 2*l1 + l2 + 2) >> 2;
+ SRC(0,1)=SRC(2,2)=SRC(4,3)=SRC(6,4)= (l0 + l1 + 1) >> 1;
+ SRC(1,1)=SRC(3,2)=SRC(5,3)=SRC(7,4)= (lt + 2*l0 + l1 + 2) >> 2;
+ SRC(0,0)=SRC(2,1)=SRC(4,2)=SRC(6,3)= (lt + l0 + 1) >> 1;
+ SRC(1,0)=SRC(3,1)=SRC(5,2)=SRC(7,3)= (l0 + 2*lt + t0 + 2) >> 2;
+ SRC(2,0)=SRC(4,1)=SRC(6,2)= (t1 + 2*t0 + lt + 2) >> 2;
+ SRC(3,0)=SRC(5,1)=SRC(7,2)= (t2 + 2*t1 + t0 + 2) >> 2;
+ SRC(4,0)=SRC(6,1)= (t3 + 2*t2 + t1 + 2) >> 2;
+ SRC(5,0)=SRC(7,1)= (t4 + 2*t3 + t2 + 2) >> 2;
+ SRC(6,0)= (t5 + 2*t4 + t3 + 2) >> 2;
+ SRC(7,0)= (t6 + 2*t5 + t4 + 2) >> 2;
+}
+static void pred8x8l_vertical_left_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_TOP;
+ PREDICT_8x8_LOAD_TOPRIGHT;
+ SRC(0,0)= (t0 + t1 + 1) >> 1;
+ SRC(0,1)= (t0 + 2*t1 + t2 + 2) >> 2;
+ SRC(0,2)=SRC(1,0)= (t1 + t2 + 1) >> 1;
+ SRC(0,3)=SRC(1,1)= (t1 + 2*t2 + t3 + 2) >> 2;
+ SRC(0,4)=SRC(1,2)=SRC(2,0)= (t2 + t3 + 1) >> 1;
+ SRC(0,5)=SRC(1,3)=SRC(2,1)= (t2 + 2*t3 + t4 + 2) >> 2;
+ SRC(0,6)=SRC(1,4)=SRC(2,2)=SRC(3,0)= (t3 + t4 + 1) >> 1;
+ SRC(0,7)=SRC(1,5)=SRC(2,3)=SRC(3,1)= (t3 + 2*t4 + t5 + 2) >> 2;
+ SRC(1,6)=SRC(2,4)=SRC(3,2)=SRC(4,0)= (t4 + t5 + 1) >> 1;
+ SRC(1,7)=SRC(2,5)=SRC(3,3)=SRC(4,1)= (t4 + 2*t5 + t6 + 2) >> 2;
+ SRC(2,6)=SRC(3,4)=SRC(4,2)=SRC(5,0)= (t5 + t6 + 1) >> 1;
+ SRC(2,7)=SRC(3,5)=SRC(4,3)=SRC(5,1)= (t5 + 2*t6 + t7 + 2) >> 2;
+ SRC(3,6)=SRC(4,4)=SRC(5,2)=SRC(6,0)= (t6 + t7 + 1) >> 1;
+ SRC(3,7)=SRC(4,5)=SRC(5,3)=SRC(6,1)= (t6 + 2*t7 + t8 + 2) >> 2;
+ SRC(4,6)=SRC(5,4)=SRC(6,2)=SRC(7,0)= (t7 + t8 + 1) >> 1;
+ SRC(4,7)=SRC(5,5)=SRC(6,3)=SRC(7,1)= (t7 + 2*t8 + t9 + 2) >> 2;
+ SRC(5,6)=SRC(6,4)=SRC(7,2)= (t8 + t9 + 1) >> 1;
+ SRC(5,7)=SRC(6,5)=SRC(7,3)= (t8 + 2*t9 + t10 + 2) >> 2;
+ SRC(6,6)=SRC(7,4)= (t9 + t10 + 1) >> 1;
+ SRC(6,7)=SRC(7,5)= (t9 + 2*t10 + t11 + 2) >> 2;
+ SRC(7,6)= (t10 + t11 + 1) >> 1;
+ SRC(7,7)= (t10 + 2*t11 + t12 + 2) >> 2;
+}
+static void pred8x8l_horizontal_up_c(uint8_t *src, int has_topleft, int has_topright, int stride)
+{
+ PREDICT_8x8_LOAD_LEFT;
+ SRC(0,0)= (l0 + l1 + 1) >> 1;
+ SRC(1,0)= (l0 + 2*l1 + l2 + 2) >> 2;
+ SRC(0,1)=SRC(2,0)= (l1 + l2 + 1) >> 1;
+ SRC(1,1)=SRC(3,0)= (l1 + 2*l2 + l3 + 2) >> 2;
+ SRC(0,2)=SRC(2,1)=SRC(4,0)= (l2 + l3 + 1) >> 1;
+ SRC(1,2)=SRC(3,1)=SRC(5,0)= (l2 + 2*l3 + l4 + 2) >> 2;
+ SRC(0,3)=SRC(2,2)=SRC(4,1)=SRC(6,0)= (l3 + l4 + 1) >> 1;
+ SRC(1,3)=SRC(3,2)=SRC(5,1)=SRC(7,0)= (l3 + 2*l4 + l5 + 2) >> 2;
+ SRC(0,4)=SRC(2,3)=SRC(4,2)=SRC(6,1)= (l4 + l5 + 1) >> 1;
+ SRC(1,4)=SRC(3,3)=SRC(5,2)=SRC(7,1)= (l4 + 2*l5 + l6 + 2) >> 2;
+ SRC(0,5)=SRC(2,4)=SRC(4,3)=SRC(6,2)= (l5 + l6 + 1) >> 1;
+ SRC(1,5)=SRC(3,4)=SRC(5,3)=SRC(7,2)= (l5 + 2*l6 + l7 + 2) >> 2;
+ SRC(0,6)=SRC(2,5)=SRC(4,4)=SRC(6,3)= (l6 + l7 + 1) >> 1;
+ SRC(1,6)=SRC(3,5)=SRC(5,4)=SRC(7,3)= (l6 + 3*l7 + 2) >> 2;
+ SRC(0,7)=SRC(1,7)=SRC(2,6)=SRC(2,7)=SRC(3,6)=
+ SRC(3,7)=SRC(4,5)=SRC(4,6)=SRC(4,7)=SRC(5,5)=
+ SRC(5,6)=SRC(5,7)=SRC(6,4)=SRC(6,5)=SRC(6,6)=
+ SRC(6,7)=SRC(7,4)=SRC(7,5)=SRC(7,6)=SRC(7,7)= l7;
+}
+#undef PREDICT_8x8_LOAD_LEFT
+#undef PREDICT_8x8_LOAD_TOP
+#undef PREDICT_8x8_LOAD_TOPLEFT
+#undef PREDICT_8x8_LOAD_TOPRIGHT
+#undef PREDICT_8x8_DC
+#undef PTR
+#undef PT
+#undef PL
+#undef SRC
+