/* FIXME how to setup second coeffient for mask tex coord */

/* 
   g3 (v0) { u0, v0, 1.0, 1.0 }  ==> {u0, v0, 1.0, 1.0, mu0, mv0, 1.0, 1.0}  Co[0](u0) Co[1](v0) Co[2](mu0) Co[3](mv0)
   g4 (v1) { u1, v1, 1.0, 1.0 }  ==> {u1, v1, 1.0, 1.0, mu1, mv1, 1.0, 1.0}
   g5 (v2) { u2, v2 }  ==> (u2, v2, mu2, mv2}
   g6      { 1/(x1-x0), 1/(y1-y0) }
   g7      { u1-u0, v1-v0, 0, 0}  ==>{u1-u0, v1-v0,0, 0, mu1-mu0, mv1-mv0, 0, 0}
	   -> { (u1-u0)/(x1-x0), (v1-v0)/(y1-y0) }  ==>{(u1-u0)/(x1-x0), (v1-v0)/(y1-y0),(mu1-mu0)/(x1-x0), (mv1-mv0)/(y1-y0)
		Cx,		 Cy 			Cx[0],		 Cy[0],		 Cx[1], 	    Cy[1]
 */

/* assign Cx[0], Cx[1] to src, same to Cy, Co 
          Cx[2], Cx[3] to mask, same to Cy, Co */

send (1) 0 g6<1>F g1.12<0,1,0>F math inv scalar mlen 1 rlen 1 { align1 };
send (1) 0 g6.4<1>F g1.20<0,1,0>F math inv scalar mlen 1 rlen 1 { align1 };
add (8) g7<1>F g4<8,8,1>F -g3<8,8,1>F { align1 };
/* Cx[0] */
mul (1) g7<1>F g7<0,1,0>F g6<0,1,0>F { align1 };
/* Cy[0] */
mul (1) g7.4<1>F g7.4<0,1,0>F g6.4<0,1,0>F { align1 };
/* Cx[2] */
mul (1) g7.8<1>F g7.8<0,1,0>F g6<0,1,0>F { align1 };
/* Cy[2] */
mul (1) g7.12<1>F g7.12<0,1,0>F g6.4<0,1,0>F { align1 };

/* src Cx[0], Cx[1] */
mov (8) m1<1>F g7<0,1,0>F { align1 };
/* mask Cx[2], Cx[3] */
mov (1) m1.8<1>F g7.8<0,1,0>F { align1 };
mov (1) m1.12<1>F g7.8<0,1,0>F { align1 };
/* src Cy[0], Cy[1] */
mov (8) m2<1>F g7.4<0,1,0>F { align1 };
/* mask Cy[2], Cy[3] */
mov (1) m2.8<1>F g7.12<0,1,0>F { align1 };
mov (1) m2.12<1>F g7.12<0,1,0>F { align1 };
/* src Co[0], Co[1] */
mov (8) m3<1>F g3<8,8,1>F { align1 };
/* mask Co[2], Co[3] */
mov (1) m3.8<1>F g3.8<0,1,0>F { align1 };
mov (1) m3.12<1>F g3.12<0,1,0>F { align1 };

send (8) 0 null g0<8,8,1>F urb 0 transpose used complete mlen 4 rlen 0 { align1 EOT };
nop;
nop;
nop;
nop;
nop;
nop;
nop;
nop;