Lines Matching refs:Float4
78 this->x = Float4(x); in Vector4f()
79 this->y = Float4(y); in Vector4f()
80 this->z = Float4(z); in Vector4f()
81 this->w = Float4(w); in Vector4f()
102 Float4 &Vector4f::operator[](int i) in operator []()
115 Float4 exponential2(RValue<Float4> x, bool pp) in exponential2()
123 Float4 x0 = x; in exponential2()
124 x0 = Min(x0, As<Float4>(Int4(0x43010000))); // 129.00000e+0f in exponential2()
125 x0 = Max(x0, As<Float4>(Int4(0xC2FDFFFF))); // -126.99999e+0f in exponential2()
127 Int4 i = RoundInt(x0 - Float4(0.5f)); in exponential2()
128 …Float4 ii = As<Float4>((i + Int4(127)) << 23); // Add single-precision bias, and shift into expo… in exponential2()
132 Float4 f = x0 - Float4(i); in exponential2()
133 Float4 ff = As<Float4>(Int4(0x3AF61905)); // 1.8775767e-3f in exponential2()
134 ff = ff * f + As<Float4>(Int4(0x3C134806)); // 8.9893397e-3f in exponential2()
135 ff = ff * f + As<Float4>(Int4(0x3D64AA23)); // 5.5826318e-2f in exponential2()
136 ff = ff * f + As<Float4>(Int4(0x3E75EAD4)); // 2.4015361e-1f in exponential2()
137 ff = ff * f + As<Float4>(Int4(0x3F31727B)); // 6.9315308e-1f in exponential2()
138 ff = ff * f + Float4(1.0f); in exponential2()
143 Float4 logarithm2(RValue<Float4> x, bool absolute, bool pp) in logarithm2()
145 Float4 x0; in logarithm2()
146 Float4 x1; in logarithm2()
147 Float4 x2; in logarithm2()
148 Float4 x3; in logarithm2()
152 x1 = As<Float4>(As<Int4>(x0) & Int4(0x7F800000)); in logarithm2()
153 x1 = As<Float4>(As<UInt4>(x1) >> 8); in logarithm2()
154 x1 = As<Float4>(As<Int4>(x1) | As<Int4>(Float4(1.0f))); in logarithm2()
155 x1 = (x1 - Float4(1.4960938f)) * Float4(256.0f); // FIXME: (x1 - 1.4960938f) * 256.0f; in logarithm2()
156 x0 = As<Float4>((As<Int4>(x0) & Int4(0x007FFFFF)) | As<Int4>(Float4(1.0f))); in logarithm2()
158 x2 = (Float4(9.5428179e-2f) * x0 + Float4(4.7779095e-1f)) * x0 + Float4(1.9782813e-1f); in logarithm2()
159 …x3 = ((Float4(1.6618466e-2f) * x0 + Float4(2.0350508e-1f)) * x0 + Float4(2.7382900e-1f)) * x0 + Fl… in logarithm2()
162 x1 += (x0 - Float4(1.0f)) * x2; in logarithm2()
165 return As<Float4>((pos_inf_x & As<Int4>(x)) | (~pos_inf_x & As<Int4>(x1))); in logarithm2()
168 Float4 exponential(RValue<Float4> x, bool pp) in exponential()
171 return exponential2(Float4(1.44269504f) * x, pp); // 1/ln(2) in exponential()
174 Float4 logarithm(RValue<Float4> x, bool absolute, bool pp) in logarithm()
177 return Float4(6.93147181e-1f) * logarithm2(x, absolute, pp); // ln(2) in logarithm()
180 Float4 power(RValue<Float4> x, RValue<Float4> y, bool pp) in power()
182 Float4 log = logarithm2(x, true, pp); in power()
187 Float4 reciprocal(RValue<Float4> x, bool pp, bool finite, bool exactAtPow2) in reciprocal()
189 Float4 rcp; in reciprocal()
193 rcp = Float4(1.0f) / x; in reciprocal()
208 rcp = Min(rcp, Float4((float&)big)); in reciprocal()
214 Float4 reciprocalSquareRoot(RValue<Float4> x, bool absolute, bool pp) in reciprocalSquareRoot()
216 Float4 abs = x; in reciprocalSquareRoot()
223 Float4 rsq; in reciprocalSquareRoot()
227 rsq = Float4(1.0f) / Sqrt(abs); in reciprocalSquareRoot()
235 rsq = rsq * (Float4(3.0f) - rsq * rsq * abs) * Float4(0.5f); in reciprocalSquareRoot()
238 rsq = As<Float4>(CmpNEQ(As<Int4>(abs), Int4(0x7F800000)) & As<Int4>(rsq)); in reciprocalSquareRoot()
244 Float4 modulo(RValue<Float4> x, RValue<Float4> y) in modulo()
249 Float4 sine_pi(RValue<Float4> x, bool pp) in sine_pi()
251 const Float4 A = Float4(-4.05284734e-1f); // -4/pi^2 in sine_pi()
252 const Float4 B = Float4(1.27323954e+0f); // 4/pi in sine_pi()
253 const Float4 C = Float4(7.75160950e-1f); in sine_pi()
254 const Float4 D = Float4(2.24839049e-1f); in sine_pi()
257 Float4 sin = x * (Abs(x) * A + B); in sine_pi()
268 Float4 cosine_pi(RValue<Float4> x, bool pp) in cosine_pi()
271 Float4 y = x + Float4(1.57079632e+0f); in cosine_pi()
274 y -= As<Float4>(CmpNLT(y, Float4(3.14159265e+0f)) & As<Int4>(Float4(6.28318530e+0f))); in cosine_pi()
279 Float4 sine(RValue<Float4> x, bool pp) in sine()
282 Float4 y = x * Float4(1.59154943e-1f); // 1/2pi in sine()
292 Float4 y2 = y * y; in sine()
293 …Float4 c1 = y2 * (y2 * (y2 * Float4(-0.0204391631f) + Float4(0.2536086171f)) + Float4(-1.233697792… in sine()
294 …Float4 s1 = y * (y2 * (y2 * (y2 * Float4(-0.0046075748f) + Float4(0.0796819754f)) + Float4(-0.6459… in sine()
295 Float4 c2 = (c1 * c1) - (s1 * s1); in sine()
296 Float4 s2 = Float4(2.0f) * s1 * c1; in sine()
297 return Float4(2.0f) * s2 * c2 * reciprocal(s2 * s2 + c2 * c2, pp, true); in sine()
300 const Float4 A = Float4(-16.0f); in sine()
301 const Float4 B = Float4(8.0f); in sine()
302 const Float4 C = Float4(7.75160950e-1f); in sine()
303 const Float4 D = Float4(2.24839049e-1f); in sine()
306 Float4 sin = y * (Abs(y) * A + B); in sine()
317 Float4 cosine(RValue<Float4> x, bool pp) in cosine()
320 Float4 y = x + Float4(1.57079632e+0f); in cosine()
324 Float4 tangent(RValue<Float4> x, bool pp) in tangent()
329 Float4 arccos(RValue<Float4> x, bool pp) in arccos()
332 return Float4(1.57079632e+0f) - arcsin(x); in arccos()
335 Float4 arcsin(RValue<Float4> x, bool pp) in arcsin()
340 return x * (Float4(1.57079632e+0f) - Sqrt(Float4(1.0f) - x*x) * Float4(6.28318531e-1f)); in arcsin()
345 const Float4 half_pi(1.57079632f); in arcsin()
346 const Float4 a0(1.5707288f); in arcsin()
347 const Float4 a1(-0.2121144f); in arcsin()
348 const Float4 a2(0.0742610f); in arcsin()
349 const Float4 a3(-0.0187293f); in arcsin()
350 Float4 absx = Abs(x); in arcsin()
351 …return As<Float4>(As<Int4>(half_pi - Sqrt(Float4(1.0f) - absx) * (a0 + absx * (a1 + absx * (a2 + a… in arcsin()
357 Float4 arctan_01(Float4 x, bool pp) in arctan_01()
361 return x * (Float4(-0.27f) * x + Float4(1.05539816f)); in arctan_01()
366 const Float4 a2(-0.3333314528f); in arctan_01()
367 const Float4 a4(0.1999355085f); in arctan_01()
368 const Float4 a6(-0.1420889944f); in arctan_01()
369 const Float4 a8(0.1065626393f); in arctan_01()
370 const Float4 a10(-0.0752896400f); in arctan_01()
371 const Float4 a12(0.0429096138f); in arctan_01()
372 const Float4 a14(-0.0161657367f); in arctan_01()
373 const Float4 a16(0.0028662257f); in arctan_01()
374 Float4 x2 = x * x; in arctan_01()
379 Float4 arctan(RValue<Float4> x, bool pp) in arctan()
381 Float4 absx = Abs(x); in arctan()
382 Int4 O = CmpNLT(absx, Float4(1.0f)); in arctan()
383 …Float4 y = As<Float4>((O & As<Int4>(Float4(1.0f) / absx)) | (~O & As<Int4>(absx))); // FIXME: Vect… in arctan()
385 const Float4 half_pi(1.57079632f); in arctan()
386 Float4 theta = arctan_01(y, pp); in arctan()
387 …return As<Float4>(((O & As<Int4>(half_pi - theta)) | (~O & As<Int4>(theta))) ^ // FIXME: Vector se… in arctan()
391 Float4 arctan(RValue<Float4> y, RValue<Float4> x, bool pp) in arctan()
393 const Float4 pi(3.14159265f); // pi in arctan()
394 const Float4 minus_pi(-3.14159265f); // -pi in arctan()
395 const Float4 half_pi(1.57079632f); // pi/2 in arctan()
396 const Float4 quarter_pi(7.85398163e-1f); // pi/4 in arctan()
399 Int4 S = CmpLT(y, Float4(0.0f)); in arctan()
400 Float4 theta = As<Float4>(S & As<Int4>(minus_pi)); in arctan()
401 Float4 x0 = As<Float4>((As<Int4>(y) & Int4(0x80000000)) ^ As<Int4>(x)); in arctan()
402 Float4 y0 = Abs(y); in arctan()
405 Int4 Q = CmpLT(x0, Float4(0.0f)); in arctan()
406 theta += As<Float4>(Q & As<Int4>(half_pi)); in arctan()
407 Float4 x1 = As<Float4>((Q & As<Int4>(y0)) | (~Q & As<Int4>(x0))); // FIXME: Vector select in arctan()
408 Float4 y1 = As<Float4>((Q & As<Int4>(-x0)) | (~Q & As<Int4>(y0))); // FIXME: Vector select in arctan()
412 Float4 x2 = As<Float4>((O & As<Int4>(y1)) | (~O & As<Int4>(x1))); // FIXME: Vector select in arctan()
413 Float4 y2 = As<Float4>((O & As<Int4>(x1)) | (~O & As<Int4>(y1))); // FIXME: Vector select in arctan()
416 Int4 zero_x = CmpEQ(x2, Float4(0.0f)); in arctan()
418 Float4 atan2_theta = arctan_01(y2 / x2, pp); in arctan()
419 …theta += As<Float4>((~zero_x & ~inf_y & ((O & As<Int4>(half_pi - atan2_theta)) | (~O & (As<Int4>(a… in arctan()
424 …return As<Float4>((precision_loss & As<Int4>(-atan2_theta)) | (~precision_loss & As<Int4>(theta)))… in arctan()
427 Float4 sineh(RValue<Float4> x, bool pp) in sineh()
429 return (exponential(x, pp) - exponential(-x, pp)) * Float4(0.5f); in sineh()
432 Float4 cosineh(RValue<Float4> x, bool pp) in cosineh()
434 return (exponential(x, pp) + exponential(-x, pp)) * Float4(0.5f); in cosineh()
437 Float4 tangenth(RValue<Float4> x, bool pp) in tangenth()
439 Float4 e_x = exponential(x, pp); in tangenth()
440 Float4 e_minus_x = exponential(-x, pp); in tangenth()
444 Float4 arccosh(RValue<Float4> x, bool pp) in arccosh()
446 return logarithm(x + Sqrt(x + Float4(1.0f)) * Sqrt(x - Float4(1.0f)), pp); in arccosh()
449 Float4 arcsinh(RValue<Float4> x, bool pp) in arcsinh()
451 return logarithm(x + Sqrt(x * x + Float4(1.0f)), pp); in arcsinh()
454 Float4 arctanh(RValue<Float4> x, bool pp) in arctanh()
456 return logarithm((Float4(1.0f) + x) / (Float4(1.0f) - x), pp) * Float4(0.5f); in arctanh()
459 Float4 dot2(const Vector4f &v0, const Vector4f &v1) in dot2()
464 Float4 dot3(const Vector4f &v0, const Vector4f &v1) in dot3()
469 Float4 dot4(const Vector4f &v0, const Vector4f &v1) in dot4()
499 void transpose4x4(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3) in transpose4x4()
501 Float4 tmp0 = UnpackLow(row0, row1); in transpose4x4()
502 Float4 tmp1 = UnpackLow(row2, row3); in transpose4x4()
503 Float4 tmp2 = UnpackHigh(row0, row1); in transpose4x4()
504 Float4 tmp3 = UnpackHigh(row2, row3); in transpose4x4()
506 row0 = Float4(tmp0.xy, tmp1.xy); in transpose4x4()
507 row1 = Float4(tmp0.zw, tmp1.zw); in transpose4x4()
508 row2 = Float4(tmp2.xy, tmp3.xy); in transpose4x4()
509 row3 = Float4(tmp2.zw, tmp3.zw); in transpose4x4()
512 void transpose4x3(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3) in transpose4x3()
514 Float4 tmp0 = UnpackLow(row0, row1); in transpose4x3()
515 Float4 tmp1 = UnpackLow(row2, row3); in transpose4x3()
516 Float4 tmp2 = UnpackHigh(row0, row1); in transpose4x3()
517 Float4 tmp3 = UnpackHigh(row2, row3); in transpose4x3()
519 row0 = Float4(tmp0.xy, tmp1.xy); in transpose4x3()
520 row1 = Float4(tmp0.zw, tmp1.zw); in transpose4x3()
521 row2 = Float4(tmp2.xy, tmp3.xy); in transpose4x3()
524 void transpose4x2(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3) in transpose4x2()
526 Float4 tmp0 = UnpackLow(row0, row1); in transpose4x2()
527 Float4 tmp1 = UnpackLow(row2, row3); in transpose4x2()
529 row0 = Float4(tmp0.xy, tmp1.xy); in transpose4x2()
530 row1 = Float4(tmp0.zw, tmp1.zw); in transpose4x2()
533 void transpose4x1(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3) in transpose4x1()
535 Float4 tmp0 = UnpackLow(row0, row1); in transpose4x1()
536 Float4 tmp1 = UnpackLow(row2, row3); in transpose4x1()
538 row0 = Float4(tmp0.xy, tmp1.xy); in transpose4x1()
541 void transpose2x4(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3) in transpose2x4()
543 Float4 tmp01 = UnpackLow(row0, row1); in transpose2x4()
544 Float4 tmp23 = UnpackHigh(row0, row1); in transpose2x4()
547 row1 = Float4(tmp01.zw, row1.zw); in transpose2x4()
549 row3 = Float4(tmp23.zw, row3.zw); in transpose2x4()
552 void transpose4xN(Float4 &row0, Float4 &row1, Float4 &row2, Float4 &row3, int N) in transpose4xN()
597 void RegisterFile::scatter_x(Int4 index, RValue<Float4> r) in scatter_x()
612 void RegisterFile::scatter_y(Int4 index, RValue<Float4> r) in scatter_y()
627 void RegisterFile::scatter_z(Int4 index, RValue<Float4> r) in scatter_z()
642 void RegisterFile::scatter_w(Int4 index, RValue<Float4> r) in scatter_w()
661 dst.x = As<Float4>(RoundInt(src.x)); in mov()
662 dst.y = As<Float4>(RoundInt(src.y)); in mov()
663 dst.z = As<Float4>(RoundInt(src.z)); in mov()
664 dst.w = As<Float4>(RoundInt(src.w)); in mov()
682 dst.x = As<Float4>(-As<Int4>(src.x)); in ineg()
683 dst.y = As<Float4>(-As<Int4>(src.y)); in ineg()
684 dst.z = As<Float4>(-As<Int4>(src.z)); in ineg()
685 dst.w = As<Float4>(-As<Int4>(src.w)); in ineg()
690 dst.x = As<Float4>(CmpNEQ(src.x, Float4(0.0f))); in f2b()
691 dst.y = As<Float4>(CmpNEQ(src.y, Float4(0.0f))); in f2b()
692 dst.z = As<Float4>(CmpNEQ(src.z, Float4(0.0f))); in f2b()
693 dst.w = As<Float4>(CmpNEQ(src.w, Float4(0.0f))); in f2b()
698 dst.x = As<Float4>(As<Int4>(src.x) & As<Int4>(Float4(1.0f))); in b2f()
699 dst.y = As<Float4>(As<Int4>(src.y) & As<Int4>(Float4(1.0f))); in b2f()
700 dst.z = As<Float4>(As<Int4>(src.z) & As<Int4>(Float4(1.0f))); in b2f()
701 dst.w = As<Float4>(As<Int4>(src.w) & As<Int4>(Float4(1.0f))); in b2f()
706 dst.x = As<Float4>(Int4(src.x)); in f2i()
707 dst.y = As<Float4>(Int4(src.y)); in f2i()
708 dst.z = As<Float4>(Int4(src.z)); in f2i()
709 dst.w = As<Float4>(Int4(src.w)); in f2i()
714 dst.x = Float4(As<Int4>(src.x)); in i2f()
715 dst.y = Float4(As<Int4>(src.y)); in i2f()
716 dst.z = Float4(As<Int4>(src.z)); in i2f()
717 dst.w = Float4(As<Int4>(src.w)); in i2f()
722 dst.x = As<Float4>(UInt4(src.x)); in f2u()
723 dst.y = As<Float4>(UInt4(src.y)); in f2u()
724 dst.z = As<Float4>(UInt4(src.z)); in f2u()
725 dst.w = As<Float4>(UInt4(src.w)); in f2u()
730 dst.x = Float4(As<UInt4>(src.x)); in u2f()
731 dst.y = Float4(As<UInt4>(src.y)); in u2f()
732 dst.z = Float4(As<UInt4>(src.z)); in u2f()
733 dst.w = Float4(As<UInt4>(src.w)); in u2f()
738 dst.x = As<Float4>(CmpNEQ(As<Int4>(src.x), Int4(0))); in i2b()
739 dst.y = As<Float4>(CmpNEQ(As<Int4>(src.y), Int4(0))); in i2b()
740 dst.z = As<Float4>(CmpNEQ(As<Int4>(src.z), Int4(0))); in i2b()
741 dst.w = As<Float4>(CmpNEQ(As<Int4>(src.w), Int4(0))); in i2b()
746 dst.x = As<Float4>(As<Int4>(src.x) & Int4(1)); in b2i()
747 dst.y = As<Float4>(As<Int4>(src.y) & Int4(1)); in b2i()
748 dst.z = As<Float4>(As<Int4>(src.z) & Int4(1)); in b2i()
749 dst.w = As<Float4>(As<Int4>(src.w) & Int4(1)); in b2i()
762 dst.x = As<Float4>(As<Int4>(src0.x) + As<Int4>(src1.x)); in iadd()
763 dst.y = As<Float4>(As<Int4>(src0.y) + As<Int4>(src1.y)); in iadd()
764 dst.z = As<Float4>(As<Int4>(src0.z) + As<Int4>(src1.z)); in iadd()
765 dst.w = As<Float4>(As<Int4>(src0.w) + As<Int4>(src1.w)); in iadd()
778 dst.x = As<Float4>(As<Int4>(src0.x) - As<Int4>(src1.x)); in isub()
779 dst.y = As<Float4>(As<Int4>(src0.y) - As<Int4>(src1.y)); in isub()
780 dst.z = As<Float4>(As<Int4>(src0.z) - As<Int4>(src1.z)); in isub()
781 dst.w = As<Float4>(As<Int4>(src0.w) - As<Int4>(src1.w)); in isub()
794 dst.x = As<Float4>(As<Int4>(src0.x) * As<Int4>(src1.x) + As<Int4>(src2.x)); in imad()
795 dst.y = As<Float4>(As<Int4>(src0.y) * As<Int4>(src1.y) + As<Int4>(src2.y)); in imad()
796 dst.z = As<Float4>(As<Int4>(src0.z) * As<Int4>(src1.z) + As<Int4>(src2.z)); in imad()
797 dst.w = As<Float4>(As<Int4>(src0.w) * As<Int4>(src1.w) + As<Int4>(src2.w)); in imad()
810 dst.x = As<Float4>(As<Int4>(src0.x) * As<Int4>(src1.x)); in imul()
811 dst.y = As<Float4>(As<Int4>(src0.y) * As<Int4>(src1.y)); in imul()
812 dst.z = As<Float4>(As<Int4>(src0.z) * As<Int4>(src1.z)); in imul()
813 dst.w = As<Float4>(As<Int4>(src0.w) * As<Int4>(src1.w)); in imul()
818 Float4 rcp = reciprocal(src.x, pp, true, true); in rcpx()
836 Float4 intMax(As<Float4>(Int4(INT_MAX))); in idiv()
838 dst.x = As<Float4>(As<Int4>(src0.x) / As<Int4>(dst.x)); in idiv()
840 dst.y = As<Float4>(As<Int4>(src0.y) / As<Int4>(dst.y)); in idiv()
842 dst.z = As<Float4>(As<Int4>(src0.z) / As<Int4>(dst.z)); in idiv()
844 dst.w = As<Float4>(As<Int4>(src0.w) / As<Int4>(dst.w)); in idiv()
849 Float4 uintMax(As<Float4>(UInt4(UINT_MAX))); in udiv()
851 dst.x = As<Float4>(As<UInt4>(src0.x) / As<UInt4>(dst.x)); in udiv()
853 dst.y = As<Float4>(As<UInt4>(src0.y) / As<UInt4>(dst.y)); in udiv()
855 dst.z = As<Float4>(As<UInt4>(src0.z) / As<UInt4>(dst.z)); in udiv()
857 dst.w = As<Float4>(As<UInt4>(src0.w) / As<UInt4>(dst.w)); in udiv()
870 Float4 intMax(As<Float4>(Int4(INT_MAX))); in imod()
872 dst.x = As<Float4>(As<Int4>(src0.x) % As<Int4>(dst.x)); in imod()
874 dst.y = As<Float4>(As<Int4>(src0.y) % As<Int4>(dst.y)); in imod()
876 dst.z = As<Float4>(As<Int4>(src0.z) % As<Int4>(dst.z)); in imod()
878 dst.w = As<Float4>(As<Int4>(src0.w) % As<Int4>(dst.w)); in imod()
883 Float4 uintMax(As<Float4>(UInt4(UINT_MAX))); in umod()
885 dst.x = As<Float4>(As<UInt4>(src0.x) % As<UInt4>(dst.x)); in umod()
887 dst.y = As<Float4>(As<UInt4>(src0.y) % As<UInt4>(dst.y)); in umod()
889 dst.z = As<Float4>(As<UInt4>(src0.z) % As<UInt4>(dst.z)); in umod()
891 dst.w = As<Float4>(As<UInt4>(src0.w) % As<UInt4>(dst.w)); in umod()
896 dst.x = As<Float4>(As<Int4>(src0.x) << As<Int4>(src1.x)); in shl()
897 dst.y = As<Float4>(As<Int4>(src0.y) << As<Int4>(src1.y)); in shl()
898 dst.z = As<Float4>(As<Int4>(src0.z) << As<Int4>(src1.z)); in shl()
899 dst.w = As<Float4>(As<Int4>(src0.w) << As<Int4>(src1.w)); in shl()
904 dst.x = As<Float4>(As<Int4>(src0.x) >> As<Int4>(src1.x)); in ishr()
905 dst.y = As<Float4>(As<Int4>(src0.y) >> As<Int4>(src1.y)); in ishr()
906 dst.z = As<Float4>(As<Int4>(src0.z) >> As<Int4>(src1.z)); in ishr()
907 dst.w = As<Float4>(As<Int4>(src0.w) >> As<Int4>(src1.w)); in ishr()
912 dst.x = As<Float4>(As<UInt4>(src0.x) >> As<UInt4>(src1.x)); in ushr()
913 dst.y = As<Float4>(As<UInt4>(src0.y) >> As<UInt4>(src1.y)); in ushr()
914 dst.z = As<Float4>(As<UInt4>(src0.z) >> As<UInt4>(src1.z)); in ushr()
915 dst.w = As<Float4>(As<UInt4>(src0.w) >> As<UInt4>(src1.w)); in ushr()
920 Float4 rsq = reciprocalSquareRoot(src.x, true, pp); in rsqx()
944 void ShaderCore::len2(Float4 &dst, const Vector4f &src, bool pp) in len2()
949 void ShaderCore::len3(Float4 &dst, const Vector4f &src, bool pp) in len3()
954 void ShaderCore::len4(Float4 &dst, const Vector4f &src, bool pp) in len4()
959 void ShaderCore::dist1(Float4 &dst, const Vector4f &src0, const Vector4f &src1, bool pp) in dist1()
964 void ShaderCore::dist2(Float4 &dst, const Vector4f &src0, const Vector4f &src1, bool pp) in dist2()
966 Float4 dx = src0.x - src1.x; in dist2()
967 Float4 dy = src0.y - src1.y; in dist2()
968 Float4 dot2 = dx * dx + dy * dy; in dist2()
972 void ShaderCore::dist3(Float4 &dst, const Vector4f &src0, const Vector4f &src1, bool pp) in dist3()
974 Float4 dx = src0.x - src1.x; in dist3()
975 Float4 dy = src0.y - src1.y; in dist3()
976 Float4 dz = src0.z - src1.z; in dist3()
977 Float4 dot3 = dx * dx + dy * dy + dz * dz; in dist3()
981 void ShaderCore::dist4(Float4 &dst, const Vector4f &src0, const Vector4f &src1, bool pp) in dist4()
983 Float4 dx = src0.x - src1.x; in dist4()
984 Float4 dy = src0.y - src1.y; in dist4()
985 Float4 dz = src0.z - src1.z; in dist4()
986 Float4 dw = src0.w - src1.w; in dist4()
987 Float4 dot4 = dx * dx + dy * dy + dz * dz + dw * dw; in dist4()
993 Float4 t = src0.x * src1.x; in dp1()
1003 Float4 t = dot2(src0, src1); in dp2()
1013 Float4 t = dot2(src0, src1) + src2.x; in dp2add()
1023 Float4 dot = dot3(src0, src1); in dp3()
1033 Float4 dot = dot4(src0, src1); in dp4()
1051 dst.x = As<Float4>(Min(As<Int4>(src0.x), As<Int4>(src1.x))); in imin()
1052 dst.y = As<Float4>(Min(As<Int4>(src0.y), As<Int4>(src1.y))); in imin()
1053 dst.z = As<Float4>(Min(As<Int4>(src0.z), As<Int4>(src1.z))); in imin()
1054 dst.w = As<Float4>(Min(As<Int4>(src0.w), As<Int4>(src1.w))); in imin()
1059 dst.x = As<Float4>(Min(As<UInt4>(src0.x), As<UInt4>(src1.x))); in umin()
1060 dst.y = As<Float4>(Min(As<UInt4>(src0.y), As<UInt4>(src1.y))); in umin()
1061 dst.z = As<Float4>(Min(As<UInt4>(src0.z), As<UInt4>(src1.z))); in umin()
1062 dst.w = As<Float4>(Min(As<UInt4>(src0.w), As<UInt4>(src1.w))); in umin()
1075 dst.x = As<Float4>(Max(As<Int4>(src0.x), As<Int4>(src1.x))); in imax()
1076 dst.y = As<Float4>(Max(As<Int4>(src0.y), As<Int4>(src1.y))); in imax()
1077 dst.z = As<Float4>(Max(As<Int4>(src0.z), As<Int4>(src1.z))); in imax()
1078 dst.w = As<Float4>(Max(As<Int4>(src0.w), As<Int4>(src1.w))); in imax()
1083 dst.x = As<Float4>(Max(As<Int4>(src0.x), As<Int4>(src1.x))); in umax()
1084 dst.y = As<Float4>(Max(As<Int4>(src0.y), As<Int4>(src1.y))); in umax()
1085 dst.z = As<Float4>(Max(As<Int4>(src0.z), As<Int4>(src1.z))); in umax()
1086 dst.w = As<Float4>(Max(As<Int4>(src0.w), As<Int4>(src1.w))); in umax()
1091 dst.x = As<Float4>(As<Int4>(CmpLT(src0.x, src1.x)) & As<Int4>(Float4(1.0f))); in slt()
1092 dst.y = As<Float4>(As<Int4>(CmpLT(src0.y, src1.y)) & As<Int4>(Float4(1.0f))); in slt()
1093 dst.z = As<Float4>(As<Int4>(CmpLT(src0.z, src1.z)) & As<Int4>(Float4(1.0f))); in slt()
1094 dst.w = As<Float4>(As<Int4>(CmpLT(src0.w, src1.w)) & As<Int4>(Float4(1.0f))); in slt()
1099 dst.x = As<Float4>(CmpNLT(x.x, edge.x) & As<Int4>(Float4(1.0f))); in step()
1100 dst.y = As<Float4>(CmpNLT(x.y, edge.y) & As<Int4>(Float4(1.0f))); in step()
1101 dst.z = As<Float4>(CmpNLT(x.z, edge.z) & As<Int4>(Float4(1.0f))); in step()
1102 dst.w = As<Float4>(CmpNLT(x.w, edge.w) & As<Int4>(Float4(1.0f))); in step()
1107 Float4 exp = exponential2(src.x, pp); in exp2x()
1133 Float4 log = logarithm2(src.x, true, pp); in log2x()
1159 dst.x = Float4(1.0f); in lit()
1160 dst.y = Max(src.x, Float4(0.0f)); in lit()
1162 Float4 pow; in lit()
1165 pow = Min(pow, Float4(127.9961f)); in lit()
1166 pow = Max(pow, Float4(-127.9961f)); in lit()
1169 dst.z = As<Float4>(As<Int4>(dst.z) & CmpNLT(src.x, Float4(0.0f))); in lit()
1170 dst.z = As<Float4>(As<Int4>(dst.z) & CmpNLT(src.y, Float4(0.0f))); in lit()
1172 dst.w = Float4(1.0f); in lit()
1194 dst.x = As<Float4>(IsInf(src.x)); in isinf()
1195 dst.y = As<Float4>(IsInf(src.y)); in isinf()
1196 dst.z = As<Float4>(IsInf(src.z)); in isinf()
1197 dst.w = As<Float4>(IsInf(src.w)); in isinf()
1202 dst.x = As<Float4>(IsNan(src.x)); in isnan()
1203 dst.y = As<Float4>(IsNan(src.y)); in isnan()
1204 dst.z = As<Float4>(IsNan(src.z)); in isnan()
1205 dst.w = As<Float4>(IsNan(src.w)); in isnan()
1210 …Float4 tx = Min(Max((x.x - edge0.x) / (edge1.x - edge0.x), Float4(0.0f)), Float4(1.0f)); dst.x = t… in smooth()
1211 …Float4 ty = Min(Max((x.y - edge0.y) / (edge1.y - edge0.y), Float4(0.0f)), Float4(1.0f)); dst.y = t… in smooth()
1212 …Float4 tz = Min(Max((x.z - edge0.z) / (edge1.z - edge0.z), Float4(0.0f)), Float4(1.0f)); dst.z = t… in smooth()
1213 …Float4 tw = Min(Max((x.w - edge0.w) / (edge1.w - edge0.w), Float4(0.0f)), Float4(1.0f)); dst.w = t… in smooth()
1216 void ShaderCore::floatToHalfBits(Float4& dst, const Float4& floatBits, bool storeInUpperBits) in floatToHalfBits()
1232 …UInt4 joined = ((((As<UInt4>(Min(As<Float4>(absf & UInt4(mask_round)) * As<Float4>(UInt4(c_magic)), in floatToHalfBits()
1233 … As<Float4>(UInt4(c_clamp))))) - UInt4(mask_round)) >> 13) & b_isnormal) | in floatToHalfBits()
1237 …dst = As<Float4>(storeInUpperBits ? As<UInt4>(dst) | ((joined << 16) | justsign) : joined | (justs… in floatToHalfBits()
1240 void ShaderCore::halfToFloatBits(Float4& dst, const Float4& halfBits) in halfToFloatBits()
1248 dst = As<Float4>(As<UInt4>(As<Float4>(expmant << 13) * As<Float4>(UInt4(magic))) | in halfToFloatBits()
1263 halfToFloatBits(dst.x, As<Float4>(As<UInt4>(s0.x) & UInt4(0x0000FFFF))); in unpackHalf2x16()
1264 halfToFloatBits(dst.y, As<Float4>((As<UInt4>(s0.x) & UInt4(0xFFFF0000)) >> 16)); in unpackHalf2x16()
1270 …d.x = As<Float4>((Int4(Round(Min(Max(s0.x, Float4(-1.0f)), Float4(1.0f)) * Float4(32767.0f))) & In… in packSnorm2x16()
1271 …((Int4(Round(Min(Max(s0.y, Float4(-1.0f)), Float4(1.0f)) * Float4(32767.0f))) & Int4(0xFFFF)) << 1… in packSnorm2x16()
1277 …d.x = As<Float4>((Int4(Round(Min(Max(s0.x, Float4(0.0f)), Float4(1.0f)) * Float4(65535.0f))) & Int… in packUnorm2x16()
1278 …((Int4(Round(Min(Max(s0.y, Float4(0.0f)), Float4(1.0f)) * Float4(65535.0f))) & Int4(0xFFFF)) << 16… in packUnorm2x16()
1284 …dst.x = Min(Max(Float4(As<Int4>((As<UInt4>(s0.x) & UInt4(0x0000FFFF)) << 16)) * Float4(1.0f / floa… in unpackSnorm2x16()
1285 …dst.y = Min(Max(Float4(As<Int4>(As<UInt4>(s0.x) & UInt4(0xFFFF0000))) * Float4(1.0f / float(0x7FFF… in unpackSnorm2x16()
1291 dst.x = Float4((As<UInt4>(s0.x) & UInt4(0x0000FFFF)) << 16) * Float4(1.0f / float(0xFFFF0000)); in unpackUnorm2x16()
1292 dst.y = Float4(As<UInt4>(s0.x) & UInt4(0xFFFF0000)) * Float4(1.0f / float(0xFFFF0000)); in unpackUnorm2x16()
1368 …dst.x += ((Float4(CmpLT(dst.x, src.x) & Int4(1)) * Float4(2.0f)) - Float4(1.0f)) * Float4(CmpEQ(Fr… in roundEven()
1369 …dst.y += ((Float4(CmpLT(dst.y, src.y) & Int4(1)) * Float4(2.0f)) - Float4(1.0f)) * Float4(CmpEQ(Fr… in roundEven()
1370 …dst.z += ((Float4(CmpLT(dst.z, src.z) & Int4(1)) * Float4(2.0f)) - Float4(1.0f)) * Float4(CmpEQ(Fr… in roundEven()
1371 …dst.w += ((Float4(CmpLT(dst.w, src.w) & Int4(1)) * Float4(2.0f)) - Float4(1.0f)) * Float4(CmpEQ(Fr… in roundEven()
1384 Float4 pow = power(src0.x, src1.x, pp); in powx()
1409 Int4 flip = CmpNLT(Nref.x * I.x, Float4(0.0f)) & Int4(0x80000000); in forward1()
1411 dst.x = As<Float4>(flip ^ As<Int4>(N.x)); in forward1()
1416 Int4 flip = CmpNLT(dot2(Nref, I), Float4(0.0f)) & Int4(0x80000000); in forward2()
1418 dst.x = As<Float4>(flip ^ As<Int4>(N.x)); in forward2()
1419 dst.y = As<Float4>(flip ^ As<Int4>(N.y)); in forward2()
1424 Int4 flip = CmpNLT(dot3(Nref, I), Float4(0.0f)) & Int4(0x80000000); in forward3()
1426 dst.x = As<Float4>(flip ^ As<Int4>(N.x)); in forward3()
1427 dst.y = As<Float4>(flip ^ As<Int4>(N.y)); in forward3()
1428 dst.z = As<Float4>(flip ^ As<Int4>(N.z)); in forward3()
1433 Int4 flip = CmpNLT(dot4(Nref, I), Float4(0.0f)) & Int4(0x80000000); in forward4()
1435 dst.x = As<Float4>(flip ^ As<Int4>(N.x)); in forward4()
1436 dst.y = As<Float4>(flip ^ As<Int4>(N.y)); in forward4()
1437 dst.z = As<Float4>(flip ^ As<Int4>(N.z)); in forward4()
1438 dst.w = As<Float4>(flip ^ As<Int4>(N.w)); in forward4()
1443 Float4 d = N.x * I.x; in reflect1()
1445 dst.x = I.x - Float4(2.0f) * d * N.x; in reflect1()
1450 Float4 d = dot2(N, I); in reflect2()
1452 dst.x = I.x - Float4(2.0f) * d * N.x; in reflect2()
1453 dst.y = I.y - Float4(2.0f) * d * N.y; in reflect2()
1458 Float4 d = dot3(N, I); in reflect3()
1460 dst.x = I.x - Float4(2.0f) * d * N.x; in reflect3()
1461 dst.y = I.y - Float4(2.0f) * d * N.y; in reflect3()
1462 dst.z = I.z - Float4(2.0f) * d * N.z; in reflect3()
1467 Float4 d = dot4(N, I); in reflect4()
1469 dst.x = I.x - Float4(2.0f) * d * N.x; in reflect4()
1470 dst.y = I.y - Float4(2.0f) * d * N.y; in reflect4()
1471 dst.z = I.z - Float4(2.0f) * d * N.z; in reflect4()
1472 dst.w = I.w - Float4(2.0f) * d * N.w; in reflect4()
1475 void ShaderCore::refract1(Vector4f &dst, const Vector4f &I, const Vector4f &N, const Float4 &eta) in refract1()
1477 Float4 d = N.x * I.x; in refract1()
1478 Float4 k = Float4(1.0f) - eta * eta * (Float4(1.0f) - d * d); in refract1()
1479 Int4 pos = CmpNLT(k, Float4(0.0f)); in refract1()
1480 Float4 t = (eta * d + Sqrt(k)); in refract1()
1482 dst.x = As<Float4>(pos & As<Int4>(eta * I.x - t * N.x)); in refract1()
1485 void ShaderCore::refract2(Vector4f &dst, const Vector4f &I, const Vector4f &N, const Float4 &eta) in refract2()
1487 Float4 d = dot2(N, I); in refract2()
1488 Float4 k = Float4(1.0f) - eta * eta * (Float4(1.0f) - d * d); in refract2()
1489 Int4 pos = CmpNLT(k, Float4(0.0f)); in refract2()
1490 Float4 t = (eta * d + Sqrt(k)); in refract2()
1492 dst.x = As<Float4>(pos & As<Int4>(eta * I.x - t * N.x)); in refract2()
1493 dst.y = As<Float4>(pos & As<Int4>(eta * I.y - t * N.y)); in refract2()
1496 void ShaderCore::refract3(Vector4f &dst, const Vector4f &I, const Vector4f &N, const Float4 &eta) in refract3()
1498 Float4 d = dot3(N, I); in refract3()
1499 Float4 k = Float4(1.0f) - eta * eta * (Float4(1.0f) - d * d); in refract3()
1500 Int4 pos = CmpNLT(k, Float4(0.0f)); in refract3()
1501 Float4 t = (eta * d + Sqrt(k)); in refract3()
1503 dst.x = As<Float4>(pos & As<Int4>(eta * I.x - t * N.x)); in refract3()
1504 dst.y = As<Float4>(pos & As<Int4>(eta * I.y - t * N.y)); in refract3()
1505 dst.z = As<Float4>(pos & As<Int4>(eta * I.z - t * N.z)); in refract3()
1508 void ShaderCore::refract4(Vector4f &dst, const Vector4f &I, const Vector4f &N, const Float4 &eta) in refract4()
1510 Float4 d = dot4(N, I); in refract4()
1511 Float4 k = Float4(1.0f) - eta * eta * (Float4(1.0f) - d * d); in refract4()
1512 Int4 pos = CmpNLT(k, Float4(0.0f)); in refract4()
1513 Float4 t = (eta * d + Sqrt(k)); in refract4()
1515 dst.x = As<Float4>(pos & As<Int4>(eta * I.x - t * N.x)); in refract4()
1516 dst.y = As<Float4>(pos & As<Int4>(eta * I.y - t * N.y)); in refract4()
1517 dst.z = As<Float4>(pos & As<Int4>(eta * I.z - t * N.z)); in refract4()
1518 dst.w = As<Float4>(pos & As<Int4>(eta * I.w - t * N.w)); in refract4()
1547 dst.x = As<Float4>(Abs(As<Int4>(src.x))); in iabs()
1548 dst.y = As<Float4>(Abs(As<Int4>(src.y))); in iabs()
1549 dst.z = As<Float4>(Abs(As<Int4>(src.z))); in iabs()
1550 dst.w = As<Float4>(Abs(As<Int4>(src.w))); in iabs()
1555 Float4 dot = dot2(src, src); in nrm2()
1556 Float4 rsq = reciprocalSquareRoot(dot, false, pp); in nrm2()
1566 Float4 dot = dot3(src, src); in nrm3()
1567 Float4 rsq = reciprocalSquareRoot(dot, false, pp); in nrm3()
1577 Float4 dot = dot4(src, src); in nrm4()
1578 Float4 rsq = reciprocalSquareRoot(dot, false, pp); in nrm4()
1700 Float4 frc = Frac(src.x); in expp()
1701 Float4 floor = src.x - frc; in expp()
1706 dst.w = Float4(1.0f); in expp()
1718 Float4 tmp0; in logp()
1719 Float4 tmp1; in logp()
1720 Float4 t; in logp()
1728 dst.x = Float4(r); in logp()
1731 dst.y = As<Float4>((As<Int4>(tmp1) & Int4(0x007FFFFF)) | As<Int4>(Float4(1.0f))); in logp()
1761 void ShaderCore::extract(Float4 &dst, const Vector4f &src0, const Float4 &src1) in extract()
1768 …void ShaderCore::insert(Vector4f &dst, const Vector4f &src, const Float4 &element, const Float4 &i… in insert()
1776 void ShaderCore::sgn(Float4 &dst, const Float4 &src) in sgn()
1778 Int4 neg = As<Int4>(CmpLT(src, Float4(-0.0f))) & As<Int4>(Float4(-1.0f)); in sgn()
1779 Int4 pos = As<Int4>(CmpNLE(src, Float4(+0.0f))) & As<Int4>(Float4(1.0f)); in sgn()
1780 dst = As<Float4>(neg | pos); in sgn()
1783 void ShaderCore::isgn(Float4 &dst, const Float4 &src) in isgn()
1787 dst = As<Float4>(neg | pos); in isgn()
1790 void ShaderCore::cmp0(Float4 &dst, const Float4 &src0, const Float4 &src1, const Float4 &src2) in cmp0()
1792 Int4 pos = CmpLE(Float4(0.0f), src0); in cmp0()
1796 void ShaderCore::cmp0i(Float4 &dst, const Float4 &src0, const Float4 &src1, const Float4 &src2) in cmp0i()
1802 void ShaderCore::select(Float4 &dst, RValue<Int4> src0, const Float4 &src1, const Float4 &src2) in select()
1805 dst = As<Float4>((src0 & As<Int4>(src1)) | (~src0 & As<Int4>(src2))); in select()
1813 dst.x = As<Float4>(CmpNLE(src0.x, src1.x)); in cmp()
1814 dst.y = As<Float4>(CmpNLE(src0.y, src1.y)); in cmp()
1815 dst.z = As<Float4>(CmpNLE(src0.z, src1.z)); in cmp()
1816 dst.w = As<Float4>(CmpNLE(src0.w, src1.w)); in cmp()
1819 dst.x = As<Float4>(CmpEQ(src0.x, src1.x)); in cmp()
1820 dst.y = As<Float4>(CmpEQ(src0.y, src1.y)); in cmp()
1821 dst.z = As<Float4>(CmpEQ(src0.z, src1.z)); in cmp()
1822 dst.w = As<Float4>(CmpEQ(src0.w, src1.w)); in cmp()
1825 dst.x = As<Float4>(CmpNLT(src0.x, src1.x)); in cmp()
1826 dst.y = As<Float4>(CmpNLT(src0.y, src1.y)); in cmp()
1827 dst.z = As<Float4>(CmpNLT(src0.z, src1.z)); in cmp()
1828 dst.w = As<Float4>(CmpNLT(src0.w, src1.w)); in cmp()
1831 dst.x = As<Float4>(CmpLT(src0.x, src1.x)); in cmp()
1832 dst.y = As<Float4>(CmpLT(src0.y, src1.y)); in cmp()
1833 dst.z = As<Float4>(CmpLT(src0.z, src1.z)); in cmp()
1834 dst.w = As<Float4>(CmpLT(src0.w, src1.w)); in cmp()
1837 dst.x = As<Float4>(CmpNEQ(src0.x, src1.x)); in cmp()
1838 dst.y = As<Float4>(CmpNEQ(src0.y, src1.y)); in cmp()
1839 dst.z = As<Float4>(CmpNEQ(src0.z, src1.z)); in cmp()
1840 dst.w = As<Float4>(CmpNEQ(src0.w, src1.w)); in cmp()
1843 dst.x = As<Float4>(CmpLE(src0.x, src1.x)); in cmp()
1844 dst.y = As<Float4>(CmpLE(src0.y, src1.y)); in cmp()
1845 dst.z = As<Float4>(CmpLE(src0.z, src1.z)); in cmp()
1846 dst.w = As<Float4>(CmpLE(src0.w, src1.w)); in cmp()
1858 dst.x = As<Float4>(CmpNLE(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1859 dst.y = As<Float4>(CmpNLE(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1860 dst.z = As<Float4>(CmpNLE(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1861 dst.w = As<Float4>(CmpNLE(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1864 dst.x = As<Float4>(CmpEQ(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1865 dst.y = As<Float4>(CmpEQ(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1866 dst.z = As<Float4>(CmpEQ(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1867 dst.w = As<Float4>(CmpEQ(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1870 dst.x = As<Float4>(CmpNLT(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1871 dst.y = As<Float4>(CmpNLT(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1872 dst.z = As<Float4>(CmpNLT(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1873 dst.w = As<Float4>(CmpNLT(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1876 dst.x = As<Float4>(CmpLT(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1877 dst.y = As<Float4>(CmpLT(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1878 dst.z = As<Float4>(CmpLT(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1879 dst.w = As<Float4>(CmpLT(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1882 dst.x = As<Float4>(CmpNEQ(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1883 dst.y = As<Float4>(CmpNEQ(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1884 dst.z = As<Float4>(CmpNEQ(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1885 dst.w = As<Float4>(CmpNEQ(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1888 dst.x = As<Float4>(CmpLE(As<Int4>(src0.x), As<Int4>(src1.x))); in icmp()
1889 dst.y = As<Float4>(CmpLE(As<Int4>(src0.y), As<Int4>(src1.y))); in icmp()
1890 dst.z = As<Float4>(CmpLE(As<Int4>(src0.z), As<Int4>(src1.z))); in icmp()
1891 dst.w = As<Float4>(CmpLE(As<Int4>(src0.w), As<Int4>(src1.w))); in icmp()
1903 dst.x = As<Float4>(CmpNLE(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1904 dst.y = As<Float4>(CmpNLE(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1905 dst.z = As<Float4>(CmpNLE(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1906 dst.w = As<Float4>(CmpNLE(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1909 dst.x = As<Float4>(CmpEQ(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1910 dst.y = As<Float4>(CmpEQ(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1911 dst.z = As<Float4>(CmpEQ(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1912 dst.w = As<Float4>(CmpEQ(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1915 dst.x = As<Float4>(CmpNLT(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1916 dst.y = As<Float4>(CmpNLT(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1917 dst.z = As<Float4>(CmpNLT(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1918 dst.w = As<Float4>(CmpNLT(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1921 dst.x = As<Float4>(CmpLT(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1922 dst.y = As<Float4>(CmpLT(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1923 dst.z = As<Float4>(CmpLT(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1924 dst.w = As<Float4>(CmpLT(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1927 dst.x = As<Float4>(CmpNEQ(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1928 dst.y = As<Float4>(CmpNEQ(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1929 dst.z = As<Float4>(CmpNEQ(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1930 dst.w = As<Float4>(CmpNEQ(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1933 dst.x = As<Float4>(CmpLE(As<UInt4>(src0.x), As<UInt4>(src1.x))); in ucmp()
1934 dst.y = As<Float4>(CmpLE(As<UInt4>(src0.y), As<UInt4>(src1.y))); in ucmp()
1935 dst.z = As<Float4>(CmpLE(As<UInt4>(src0.z), As<UInt4>(src1.z))); in ucmp()
1936 dst.w = As<Float4>(CmpLE(As<UInt4>(src0.w), As<UInt4>(src1.w))); in ucmp()
1943 void ShaderCore::all(Float4 &dst, const Vector4f &src) in all()
1945 dst = As<Float4>(As<Int4>(src.x) & As<Int4>(src.y) & As<Int4>(src.z) & As<Int4>(src.w)); in all()
1948 void ShaderCore::any(Float4 &dst, const Vector4f &src) in any()
1950 dst = As<Float4>(As<Int4>(src.x) | As<Int4>(src.y) | As<Int4>(src.z) | As<Int4>(src.w)); in any()
1955 dst.x = As<Float4>(As<Int4>(src.x) ^ Int4(0xFFFFFFFF)); in bitwise_not()
1956 dst.y = As<Float4>(As<Int4>(src.y) ^ Int4(0xFFFFFFFF)); in bitwise_not()
1957 dst.z = As<Float4>(As<Int4>(src.z) ^ Int4(0xFFFFFFFF)); in bitwise_not()
1958 dst.w = As<Float4>(As<Int4>(src.w) ^ Int4(0xFFFFFFFF)); in bitwise_not()
1963 dst.x = As<Float4>(As<Int4>(src0.x) | As<Int4>(src1.x)); in bitwise_or()
1964 dst.y = As<Float4>(As<Int4>(src0.y) | As<Int4>(src1.y)); in bitwise_or()
1965 dst.z = As<Float4>(As<Int4>(src0.z) | As<Int4>(src1.z)); in bitwise_or()
1966 dst.w = As<Float4>(As<Int4>(src0.w) | As<Int4>(src1.w)); in bitwise_or()
1971 dst.x = As<Float4>(As<Int4>(src0.x) ^ As<Int4>(src1.x)); in bitwise_xor()
1972 dst.y = As<Float4>(As<Int4>(src0.y) ^ As<Int4>(src1.y)); in bitwise_xor()
1973 dst.z = As<Float4>(As<Int4>(src0.z) ^ As<Int4>(src1.z)); in bitwise_xor()
1974 dst.w = As<Float4>(As<Int4>(src0.w) ^ As<Int4>(src1.w)); in bitwise_xor()
1979 dst.x = As<Float4>(As<Int4>(src0.x) & As<Int4>(src1.x)); in bitwise_and()
1980 dst.y = As<Float4>(As<Int4>(src0.y) & As<Int4>(src1.y)); in bitwise_and()
1981 dst.z = As<Float4>(As<Int4>(src0.z) & As<Int4>(src1.z)); in bitwise_and()
1982 dst.w = As<Float4>(As<Int4>(src0.w) & As<Int4>(src1.w)); in bitwise_and()
1987 dst.x = As<Float4>(CmpEQ(As<UInt4>(src0.x), As<UInt4>(src1.x)) & in equal()
1998 dst.x = As<Float4>(CmpNEQ(As<UInt4>(src0.x), As<UInt4>(src1.x)) | in notEqual()