当前位置: 首页>>代码示例>>C++>>正文


C++ VMUL函数代码示例

本文整理汇总了C++中VMUL函数的典型用法代码示例。如果您正苦于以下问题:C++ VMUL函数的具体用法?C++ VMUL怎么用?C++ VMUL使用的例子?那么, 这里精选的函数代码示例或许可以为您提供帮助。


在下文中一共展示了VMUL函数的15个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的C++代码示例。

示例1: n2bv_3

static void n2bv_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, int v, int ivs, int ovs)
{
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     int i;
     const R *xi;
     R *xo;
     xi = ii;
     xo = io;
     BEGIN_SIMD();
     for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs)) {
	  V T4, T3, T5, T1, T2, T6;
	  T4 = LD(&(xi[0]), ivs, &(xi[0]));
	  T1 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
	  T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
	  T3 = VBYI(VMUL(LDK(KP866025403), VSUB(T1, T2)));
	  T5 = VADD(T1, T2);
	  ST(&(xo[0]), VADD(T4, T5), ovs, &(xo[0]));
	  T6 = VFNMS(LDK(KP500000000), T5, T4);
	  ST(&(xo[2]), VADD(T3, T6), ovs, &(xo[2]));
	  ST(&(xo[4]), VSUB(T6, T3), ovs, &(xo[0]));
     }
     END_SIMD();
}
开发者ID:OS2World,项目名称:MM-SOUND-PM123,代码行数:24,代码来源:n2bv_3.c

示例2: n1bv_16

static void n1bv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
     DVK(KP382683432, +0.382683432365089771728459984030398866761344562);
     DVK(KP923879532, +0.923879532511286756128183189396788286822416626);
     DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
     INT i;
     const R *xi;
     R *xo;
     xi = ii;
     xo = io;
     for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
	  V Tp, T13, Tu, TY, Tm, T14, Tv, TU, T7, T16, Tx, TN, Te, T17, Ty;
	  V TQ;
	  {
	       V Tn, To, TX, Ts, Tt, TW;
	       Tn = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
	       To = LD(&(xi[WS(is, 12)]), ivs, &(xi[0]));
	       TX = VADD(Tn, To);
	       Ts = LD(&(xi[0]), ivs, &(xi[0]));
	       Tt = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
	       TW = VADD(Ts, Tt);
	       Tp = VSUB(Tn, To);
	       T13 = VADD(TW, TX);
	       Tu = VSUB(Ts, Tt);
	       TY = VSUB(TW, TX);
	  }
	  {
	       V Ti, TS, Tl, TT;
	       {
		    V Tg, Th, Tj, Tk;
		    Tg = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
		    Th = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
		    Ti = VSUB(Tg, Th);
		    TS = VADD(Tg, Th);
		    Tj = LD(&(xi[WS(is, 14)]), ivs, &(xi[0]));
		    Tk = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
		    Tl = VSUB(Tj, Tk);
		    TT = VADD(Tj, Tk);
	       }
	       Tm = VMUL(LDK(KP707106781), VSUB(Ti, Tl));
	       T14 = VADD(TS, TT);
	       Tv = VMUL(LDK(KP707106781), VADD(Ti, Tl));
	       TU = VSUB(TS, TT);
	  }
	  {
	       V T3, TL, T6, TM;
	       {
		    V T1, T2, T4, T5;
		    T1 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
		    T2 = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
		    T3 = VSUB(T1, T2);
		    TL = VADD(T1, T2);
		    T4 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
		    T5 = LD(&(xi[WS(is, 13)]), ivs, &(xi[WS(is, 1)]));
		    T6 = VSUB(T4, T5);
		    TM = VADD(T4, T5);
	       }
	       T7 = VFNMS(LDK(KP382683432), T6, VMUL(LDK(KP923879532), T3));
	       T16 = VADD(TL, TM);
	       Tx = VFMA(LDK(KP382683432), T3, VMUL(LDK(KP923879532), T6));
	       TN = VSUB(TL, TM);
	  }
	  {
	       V Ta, TO, Td, TP;
	       {
		    V T8, T9, Tb, Tc;
		    T8 = LD(&(xi[WS(is, 15)]), ivs, &(xi[WS(is, 1)]));
		    T9 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
		    Ta = VSUB(T8, T9);
		    TO = VADD(T8, T9);
		    Tb = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
		    Tc = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
		    Td = VSUB(Tb, Tc);
		    TP = VADD(Tb, Tc);
	       }
	       Te = VFMA(LDK(KP923879532), Ta, VMUL(LDK(KP382683432), Td));
	       T17 = VADD(TO, TP);
	       Ty = VFNMS(LDK(KP382683432), Ta, VMUL(LDK(KP923879532), Td));
	       TQ = VSUB(TO, TP);
	  }
	  {
	       V T15, T18, T19, T1a;
	       T15 = VSUB(T13, T14);
	       T18 = VBYI(VSUB(T16, T17));
	       ST(&(xo[WS(os, 12)]), VSUB(T15, T18), ovs, &(xo[0]));
	       ST(&(xo[WS(os, 4)]), VADD(T15, T18), ovs, &(xo[0]));
	       T19 = VADD(T13, T14);
	       T1a = VADD(T16, T17);
	       ST(&(xo[WS(os, 8)]), VSUB(T19, T1a), ovs, &(xo[0]));
	       ST(&(xo[0]), VADD(T19, T1a), ovs, &(xo[0]));
	  }
	  {
	       V TV, T11, T10, T12, TR, TZ;
	       TR = VMUL(LDK(KP707106781), VSUB(TN, TQ));
	       TV = VBYI(VSUB(TR, TU));
	       T11 = VBYI(VADD(TU, TR));
	       TZ = VMUL(LDK(KP707106781), VADD(TN, TQ));
	       T10 = VSUB(TY, TZ);
	       T12 = VADD(TY, TZ);
	       ST(&(xo[WS(os, 6)]), VADD(TV, T10), ovs, &(xo[0]));
//.........这里部分代码省略.........
开发者ID:phillipstanleymarbell,项目名称:sunflower-simulator,代码行数:101,代码来源:n1bv_16.c

示例3: n1bv_15

static void n1bv_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
     DVK(KP216506350, +0.216506350946109661690930792688234045867850657);
     DVK(KP509036960, +0.509036960455127183450980863393907648510733164);
     DVK(KP823639103, +0.823639103546331925877420039278190003029660514);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     DVK(KP484122918, +0.484122918275927110647408174972799951354115213);
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     {
	  INT i;
	  const R *xi;
	  R *xo;
	  xi = ii;
	  xo = io;
	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
	       V Ti, T11, TH, Ts, TL, TM, Tz, TC, TD, TI, T12, T13, T14, T15, T16;
	       V T17, Tf, Tj, TZ, T10;
	       {
		    V TF, Tg, Th, TG;
		    TF = LD(&(xi[0]), ivs, &(xi[0]));
		    Tg = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
		    Th = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
		    TG = VADD(Tg, Th);
		    Ti = VSUB(Tg, Th);
		    T11 = VADD(TF, TG);
		    TH = VFNMS(LDK(KP500000000), TG, TF);
	       }
	       {
		    V Tm, Tn, T3, To, Tw, Tx, Td, Ty, Tp, Tq, T6, Tr, Tt, Tu, Ta;
		    V Tv, T7, Te;
		    {
			 V T1, T2, Tb, Tc;
			 Tm = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
			 T1 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
			 T2 = LD(&(xi[WS(is, 13)]), ivs, &(xi[WS(is, 1)]));
			 Tn = VADD(T1, T2);
			 T3 = VSUB(T1, T2);
			 To = VFNMS(LDK(KP500000000), Tn, Tm);
			 Tw = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
			 Tb = LD(&(xi[WS(is, 14)]), ivs, &(xi[0]));
			 Tc = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
			 Tx = VADD(Tb, Tc);
			 Td = VSUB(Tb, Tc);
			 Ty = VFNMS(LDK(KP500000000), Tx, Tw);
		    }
		    {
			 V T4, T5, T8, T9;
			 Tp = LD(&(xi[WS(is, 12)]), ivs, &(xi[0]));
			 T4 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
			 T5 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
			 Tq = VADD(T4, T5);
			 T6 = VSUB(T4, T5);
			 Tr = VFNMS(LDK(KP500000000), Tq, Tp);
			 Tt = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
			 T8 = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
			 T9 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
			 Tu = VADD(T8, T9);
			 Ta = VSUB(T8, T9);
			 Tv = VFNMS(LDK(KP500000000), Tu, Tt);
		    }
		    Ts = VSUB(To, Tr);
		    TL = VSUB(T3, T6);
		    TM = VSUB(Ta, Td);
		    Tz = VSUB(Tv, Ty);
		    TC = VADD(To, Tr);
		    TD = VADD(Tv, Ty);
		    TI = VADD(TC, TD);
		    T12 = VADD(Tm, Tn);
		    T13 = VADD(Tp, Tq);
		    T14 = VADD(T12, T13);
		    T15 = VADD(Tt, Tu);
		    T16 = VADD(Tw, Tx);
		    T17 = VADD(T15, T16);
		    T7 = VADD(T3, T6);
		    Te = VADD(Ta, Td);
		    Tf = VMUL(LDK(KP484122918), VSUB(T7, Te));
		    Tj = VADD(T7, Te);
	       }
	       TZ = VADD(TH, TI);
	       T10 = VBYI(VMUL(LDK(KP866025403), VADD(Ti, Tj)));
	       ST(&(xo[WS(os, 5)]), VSUB(TZ, T10), ovs, &(xo[WS(os, 1)]));
	       ST(&(xo[WS(os, 10)]), VADD(T10, TZ), ovs, &(xo[0]));
	       {
		    V T1a, T18, T19, T1e, T1f, T1c, T1d, T1g, T1b;
		    T1a = VMUL(LDK(KP559016994), VSUB(T14, T17));
		    T18 = VADD(T14, T17);
		    T19 = VFNMS(LDK(KP250000000), T18, T11);
		    T1c = VSUB(T12, T13);
		    T1d = VSUB(T15, T16);
		    T1e = VBYI(VFNMS(LDK(KP951056516), T1d, VMUL(LDK(KP587785252), T1c)));
		    T1f = VBYI(VFMA(LDK(KP951056516), T1c, VMUL(LDK(KP587785252), T1d)));
		    ST(&(xo[0]), VADD(T11, T18), ovs, &(xo[0]));
		    T1g = VADD(T1a, T19);
		    ST(&(xo[WS(os, 6)]), VADD(T1f, T1g), ovs, &(xo[0]));
		    ST(&(xo[WS(os, 9)]), VSUB(T1g, T1f), ovs, &(xo[WS(os, 1)]));
		    T1b = VSUB(T19, T1a);
//.........这里部分代码省略.........
开发者ID:AlexanderB1290,项目名称:scafacos,代码行数:101,代码来源:n1bv_15.c

示例4: t1fuv_7

static void t1fuv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP900968867, +0.900968867902419126236102319507445051165919162);
     DVK(KP222520933, +0.222520933956314404288902564496794759466355569);
     DVK(KP623489801, +0.623489801858733530525004884004239810632274731);
     DVK(KP781831482, +0.781831482468029808708444526674057750232334519);
     DVK(KP974927912, +0.974927912181823607018131682993931217232785801);
     DVK(KP433883739, +0.433883739117558120475768332848358754609990728);
     {
	  INT m;
	  R *x;
	  x = ri;
	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
	       V T1, Tg, Tj, T6, Ti, Tb, Tk, Tp, To;
	       T1 = LD(&(x[0]), ms, &(x[0]));
	       {
		    V Td, Tf, Tc, Te;
		    Tc = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
		    Td = BYTWJ(&(W[TWVL * 4]), Tc);
		    Te = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
		    Tf = BYTWJ(&(W[TWVL * 6]), Te);
		    Tg = VADD(Td, Tf);
		    Tj = VSUB(Tf, Td);
	       }
	       {
		    V T3, T5, T2, T4;
		    T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
		    T3 = BYTWJ(&(W[0]), T2);
		    T4 = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
		    T5 = BYTWJ(&(W[TWVL * 10]), T4);
		    T6 = VADD(T3, T5);
		    Ti = VSUB(T5, T3);
	       }
	       {
		    V T8, Ta, T7, T9;
		    T7 = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
		    T8 = BYTWJ(&(W[TWVL * 2]), T7);
		    T9 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
		    Ta = BYTWJ(&(W[TWVL * 8]), T9);
		    Tb = VADD(T8, Ta);
		    Tk = VSUB(Ta, T8);
	       }
	       ST(&(x[0]), VADD(T1, VADD(T6, VADD(Tb, Tg))), ms, &(x[0]));
	       Tp = VBYI(VFMA(LDK(KP433883739), Ti, VFNMS(LDK(KP781831482), Tk, VMUL(LDK(KP974927912), Tj))));
	       To = VFMA(LDK(KP623489801), Tb, VFNMS(LDK(KP222520933), Tg, VFNMS(LDK(KP900968867), T6, T1)));
	       ST(&(x[WS(rs, 4)]), VSUB(To, Tp), ms, &(x[0]));
	       ST(&(x[WS(rs, 3)]), VADD(To, Tp), ms, &(x[WS(rs, 1)]));
	       {
		    V Tl, Th, Tn, Tm;
		    Tl = VBYI(VFNMS(LDK(KP781831482), Tj, VFNMS(LDK(KP433883739), Tk, VMUL(LDK(KP974927912), Ti))));
		    Th = VFMA(LDK(KP623489801), Tg, VFNMS(LDK(KP900968867), Tb, VFNMS(LDK(KP222520933), T6, T1)));
		    ST(&(x[WS(rs, 5)]), VSUB(Th, Tl), ms, &(x[WS(rs, 1)]));
		    ST(&(x[WS(rs, 2)]), VADD(Th, Tl), ms, &(x[0]));
		    Tn = VBYI(VFMA(LDK(KP781831482), Ti, VFMA(LDK(KP974927912), Tk, VMUL(LDK(KP433883739), Tj))));
		    Tm = VFMA(LDK(KP623489801), T6, VFNMS(LDK(KP900968867), Tg, VFNMS(LDK(KP222520933), Tb, T1)));
		    ST(&(x[WS(rs, 6)]), VSUB(Tm, Tn), ms, &(x[0]));
		    ST(&(x[WS(rs, 1)]), VADD(Tm, Tn), ms, &(x[WS(rs, 1)]));
	       }
	  }
     }
     VLEAVE();
}
开发者ID:Aegisub,项目名称:fftw3,代码行数:62,代码来源:t1fuv_7.c

示例5: hc2cbdftv_10

static void hc2cbdftv_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     INT m;
     for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
	  V T5, TE, Ts, Tt, TC, Tz, TH, TJ, To, Tq, T2, T4, T3, T9, Tx;
	  V Tm, TB, Td, Ty, Ti, TA, T6, T8, T7, Tl, Tk, Tj, Tc, Tb, Ta;
	  V Tf, Th, Tg, TF, TG, Te, Tn;
	  T2 = LD(&(Rp[0]), ms, &(Rp[0]));
	  T3 = LD(&(Rm[WS(rs, 4)]), -ms, &(Rm[0]));
	  T4 = VCONJ(T3);
	  T5 = VSUB(T2, T4);
	  TE = VADD(T2, T4);
	  T6 = LD(&(Rp[WS(rs, 2)]), ms, &(Rp[0]));
	  T7 = LD(&(Rm[WS(rs, 2)]), -ms, &(Rm[0]));
	  T8 = VCONJ(T7);
	  T9 = VSUB(T6, T8);
	  Tx = VADD(T6, T8);
	  Tl = LD(&(Rp[WS(rs, 1)]), ms, &(Rp[WS(rs, 1)]));
	  Tj = LD(&(Rm[WS(rs, 3)]), -ms, &(Rm[WS(rs, 1)]));
	  Tk = VCONJ(Tj);
	  Tm = VSUB(Tk, Tl);
	  TB = VADD(Tk, Tl);
	  Tc = LD(&(Rp[WS(rs, 3)]), ms, &(Rp[WS(rs, 1)]));
	  Ta = LD(&(Rm[WS(rs, 1)]), -ms, &(Rm[WS(rs, 1)]));
	  Tb = VCONJ(Ta);
	  Td = VSUB(Tb, Tc);
	  Ty = VADD(Tb, Tc);
	  Tf = LD(&(Rp[WS(rs, 4)]), ms, &(Rp[0]));
	  Tg = LD(&(Rm[0]), -ms, &(Rm[0]));
	  Th = VCONJ(Tg);
	  Ti = VSUB(Tf, Th);
	  TA = VADD(Tf, Th);
	  Ts = VSUB(T9, Td);
	  Tt = VSUB(Ti, Tm);
	  TC = VSUB(TA, TB);
	  Tz = VSUB(Tx, Ty);
	  TF = VADD(Tx, Ty);
	  TG = VADD(TA, TB);
	  TH = VADD(TF, TG);
	  TJ = VMUL(LDK(KP559016994), VSUB(TF, TG));
	  Te = VADD(T9, Td);
	  Tn = VADD(Ti, Tm);
	  To = VADD(Te, Tn);
	  Tq = VMUL(LDK(KP559016994), VSUB(Te, Tn));
	  {
	       V T1c, TX, Tv, T1b, TR, T15, TL, T17, TT, T11, TW, Tu, TQ, Tr, TP;
	       V Tp, T1, T1a, TO, T14, TD, T10, TK, TZ, TI, Tw, T16, TS, TY, TM;
	       V TU, T1e, TN, T1d, T19, T13, TV, T18, T12;
	       T1c = VADD(TE, TH);
	       TW = LDW(&(W[TWVL * 8]));
	       TX = VZMULI(TW, VADD(T5, To));
	       Tu = VBYI(VFNMS(LDK(KP951056516), Tt, VMUL(LDK(KP587785252), Ts)));
	       TQ = VBYI(VFMA(LDK(KP951056516), Ts, VMUL(LDK(KP587785252), Tt)));
	       Tp = VFNMS(LDK(KP250000000), To, T5);
	       Tr = VSUB(Tp, Tq);
	       TP = VADD(Tq, Tp);
	       T1 = LDW(&(W[TWVL * 4]));
	       Tv = VZMULI(T1, VSUB(Tr, Tu));
	       T1a = LDW(&(W[0]));
	       T1b = VZMULI(T1a, VADD(TQ, TP));
	       TO = LDW(&(W[TWVL * 16]));
	       TR = VZMULI(TO, VSUB(TP, TQ));
	       T14 = LDW(&(W[TWVL * 12]));
	       T15 = VZMULI(T14, VADD(Tu, Tr));
	       TD = VBYI(VFNMS(LDK(KP951056516), TC, VMUL(LDK(KP587785252), Tz)));
	       T10 = VBYI(VFMA(LDK(KP951056516), Tz, VMUL(LDK(KP587785252), TC)));
	       TI = VFNMS(LDK(KP250000000), TH, TE);
	       TK = VSUB(TI, TJ);
	       TZ = VADD(TJ, TI);
	       Tw = LDW(&(W[TWVL * 2]));
	       TL = VZMUL(Tw, VADD(TD, TK));
	       T16 = LDW(&(W[TWVL * 10]));
	       T17 = VZMUL(T16, VADD(T10, TZ));
	       TS = LDW(&(W[TWVL * 14]));
	       TT = VZMUL(TS, VSUB(TK, TD));
	       TY = LDW(&(W[TWVL * 6]));
	       T11 = VZMUL(TY, VSUB(TZ, T10));
	       TM = VADD(Tv, TL);
	       ST(&(Rp[WS(rs, 1)]), TM, ms, &(Rp[WS(rs, 1)]));
	       TU = VADD(TR, TT);
	       ST(&(Rp[WS(rs, 4)]), TU, ms, &(Rp[0]));
	       T1e = VCONJ(VSUB(T1c, T1b));
	       ST(&(Rm[0]), T1e, -ms, &(Rm[0]));
	       TN = VCONJ(VSUB(TL, Tv));
	       ST(&(Rm[WS(rs, 1)]), TN, -ms, &(Rm[WS(rs, 1)]));
	       T1d = VADD(T1b, T1c);
	       ST(&(Rp[0]), T1d, ms, &(Rp[0]));
	       T19 = VCONJ(VSUB(T17, T15));
	       ST(&(Rm[WS(rs, 3)]), T19, -ms, &(Rm[WS(rs, 1)]));
	       T13 = VCONJ(VSUB(T11, TX));
	       ST(&(Rm[WS(rs, 2)]), T13, -ms, &(Rm[0]));
	       TV = VCONJ(VSUB(TT, TR));
	       ST(&(Rm[WS(rs, 4)]), TV, -ms, &(Rm[0]));
	       T18 = VADD(T15, T17);
	       ST(&(Rp[WS(rs, 3)]), T18, ms, &(Rp[WS(rs, 1)]));
	       T12 = VADD(TX, T11);
//.........这里部分代码省略.........
开发者ID:phillipstanleymarbell,项目名称:sunflower-simulator,代码行数:101,代码来源:hc2cbdftv_10.c

示例6: t3bv_10

static void t3bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     {
	  INT m;
	  R *x;
	  x = ii;
	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(10, rs)) {
	       V T1, T2, T3, Ti, T6, T7, TA, Tb, To;
	       T1 = LDW(&(W[0]));
	       T2 = LDW(&(W[TWVL * 2]));
	       T3 = VZMULJ(T1, T2);
	       Ti = VZMUL(T1, T2);
	       T6 = LDW(&(W[TWVL * 4]));
	       T7 = VZMULJ(T3, T6);
	       TA = VZMULJ(Ti, T6);
	       Tb = VZMULJ(T1, T6);
	       To = VZMULJ(T2, T6);
	       {
		    V TD, TQ, Tn, Tt, Tx, TM, TN, TS, Ta, Tg, Tw, TJ, TK, TR, Tz;
		    V TC, TB;
		    Tz = LD(&(x[0]), ms, &(x[0]));
		    TB = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
		    TC = VZMUL(TA, TB);
		    TD = VSUB(Tz, TC);
		    TQ = VADD(Tz, TC);
		    {
			 V Tk, Ts, Tm, Tq;
			 {
			      V Tj, Tr, Tl, Tp;
			      Tj = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
			      Tk = VZMUL(Ti, Tj);
			      Tr = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
			      Ts = VZMUL(T1, Tr);
			      Tl = LD(&(x[WS(rs, 9)]), ms, &(x[WS(rs, 1)]));
			      Tm = VZMUL(T6, Tl);
			      Tp = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
			      Tq = VZMUL(To, Tp);
			 }
			 Tn = VSUB(Tk, Tm);
			 Tt = VSUB(Tq, Ts);
			 Tx = VADD(Tn, Tt);
			 TM = VADD(Tk, Tm);
			 TN = VADD(Tq, Ts);
			 TS = VADD(TM, TN);
		    }
		    {
			 V T5, Tf, T9, Td;
			 {
			      V T4, Te, T8, Tc;
			      T4 = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
			      T5 = VZMUL(T3, T4);
			      Te = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
			      Tf = VZMUL(T2, Te);
			      T8 = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
			      T9 = VZMUL(T7, T8);
			      Tc = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
			      Td = VZMUL(Tb, Tc);
			 }
			 Ta = VSUB(T5, T9);
			 Tg = VSUB(Td, Tf);
			 Tw = VADD(Ta, Tg);
			 TJ = VADD(T5, T9);
			 TK = VADD(Td, Tf);
			 TR = VADD(TJ, TK);
		    }
		    {
			 V Ty, TE, TF, Tv, TI, Th, Tu, TH, TG;
			 Ty = VMUL(LDK(KP559016994), VSUB(Tw, Tx));
			 TE = VADD(Tw, Tx);
			 TF = VFNMS(LDK(KP250000000), TE, TD);
			 Th = VSUB(Ta, Tg);
			 Tu = VSUB(Tn, Tt);
			 Tv = VBYI(VFMA(LDK(KP951056516), Th, VMUL(LDK(KP587785252), Tu)));
			 TI = VBYI(VFNMS(LDK(KP951056516), Tu, VMUL(LDK(KP587785252), Th)));
			 ST(&(x[WS(rs, 5)]), VADD(TD, TE), ms, &(x[WS(rs, 1)]));
			 TH = VSUB(TF, Ty);
			 ST(&(x[WS(rs, 3)]), VSUB(TH, TI), ms, &(x[WS(rs, 1)]));
			 ST(&(x[WS(rs, 7)]), VADD(TI, TH), ms, &(x[WS(rs, 1)]));
			 TG = VADD(Ty, TF);
			 ST(&(x[WS(rs, 1)]), VADD(Tv, TG), ms, &(x[WS(rs, 1)]));
			 ST(&(x[WS(rs, 9)]), VSUB(TG, Tv), ms, &(x[WS(rs, 1)]));
		    }
		    {
			 V TV, TT, TU, TP, TY, TL, TO, TX, TW;
			 TV = VMUL(LDK(KP559016994), VSUB(TR, TS));
			 TT = VADD(TR, TS);
			 TU = VFNMS(LDK(KP250000000), TT, TQ);
			 TL = VSUB(TJ, TK);
			 TO = VSUB(TM, TN);
			 TP = VBYI(VFNMS(LDK(KP951056516), TO, VMUL(LDK(KP587785252), TL)));
			 TY = VBYI(VFMA(LDK(KP951056516), TL, VMUL(LDK(KP587785252), TO)));
			 ST(&(x[0]), VADD(TQ, TT), ms, &(x[0]));
			 TX = VADD(TV, TU);
			 ST(&(x[WS(rs, 4)]), VSUB(TX, TY), ms, &(x[0]));
			 ST(&(x[WS(rs, 6)]), VADD(TY, TX), ms, &(x[0]));
			 TW = VSUB(TU, TV);
//.........这里部分代码省略.........
开发者ID:Enny1991,项目名称:ExApp,代码行数:101,代码来源:t3bv_10.c

示例7: t2bv_10

static void t2bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP618033988, +0.618033988749894848204586834365638117720309180);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     {
	  INT m;
	  R *x;
	  x = ii;
	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
	       T1 = LD(&(x[0]), ms, &(x[0]));
	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
	       {
		    V Tg, Tn, Ti, Tl;
		    Tg = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
		    Tn = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
		    Ti = LD(&(x[WS(rs, 9)]), ms, &(x[WS(rs, 1)]));
		    Tl = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
		    {
			 V T6, T8, T5, Tc;
			 T5 = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
			 Tc = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
			 {
			      V T3, Th, To, Tj, Tm, T7;
			      T7 = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
			      T3 = BYTW(&(W[TWVL * 8]), T2);
			      Th = BYTW(&(W[TWVL * 6]), Tg);
			      To = BYTW(&(W[0]), Tn);
			      Tj = BYTW(&(W[TWVL * 16]), Ti);
			      Tm = BYTW(&(W[TWVL * 10]), Tl);
			      T6 = BYTW(&(W[TWVL * 2]), T5);
			      Td = BYTW(&(W[TWVL * 4]), Tc);
			      T8 = BYTW(&(W[TWVL * 12]), T7);
			      TA = VADD(T1, T3);
			      T4 = VSUB(T1, T3);
			      Ta = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
			      Tk = VSUB(Th, Tj);
			      TE = VADD(Th, Tj);
			      Tp = VSUB(Tm, To);
			      TF = VADD(Tm, To);
			 }
			 TB = VADD(T6, T8);
			 T9 = VSUB(T6, T8);
		    }
	       }
	       Tb = BYTW(&(W[TWVL * 14]), Ta);
	       {
		    V TL, TG, Tw, Tq, TC, Te;
		    TL = VSUB(TE, TF);
		    TG = VADD(TE, TF);
		    Tw = VSUB(Tk, Tp);
		    Tq = VADD(Tk, Tp);
		    TC = VADD(Tb, Td);
		    Te = VSUB(Tb, Td);
		    {
			 V TM, TD, Tv, Tf;
			 TM = VSUB(TB, TC);
			 TD = VADD(TB, TC);
			 Tv = VSUB(T9, Te);
			 Tf = VADD(T9, Te);
			 {
			      V TP, TN, TH, TJ, Tz, Tx, Tr, Tt, TI, Ts;
			      TP = VMUL(LDK(KP951056516), VFMA(LDK(KP618033988), TL, TM));
			      TN = VMUL(LDK(KP951056516), VFNMS(LDK(KP618033988), TM, TL));
			      TH = VADD(TD, TG);
			      TJ = VSUB(TD, TG);
			      Tz = VMUL(LDK(KP951056516), VFNMS(LDK(KP618033988), Tv, Tw));
			      Tx = VMUL(LDK(KP951056516), VFMA(LDK(KP618033988), Tw, Tv));
			      Tr = VADD(Tf, Tq);
			      Tt = VSUB(Tf, Tq);
			      ST(&(x[0]), VADD(TA, TH), ms, &(x[0]));
			      TI = VFNMS(LDK(KP250000000), TH, TA);
			      ST(&(x[WS(rs, 5)]), VADD(T4, Tr), ms, &(x[WS(rs, 1)]));
			      Ts = VFNMS(LDK(KP250000000), Tr, T4);
			      {
				   V TK, TO, Tu, Ty;
				   TK = VFNMS(LDK(KP559016994), TJ, TI);
				   TO = VFMA(LDK(KP559016994), TJ, TI);
				   Tu = VFMA(LDK(KP559016994), Tt, Ts);
				   Ty = VFNMS(LDK(KP559016994), Tt, Ts);
				   ST(&(x[WS(rs, 8)]), VFMAI(TN, TK), ms, &(x[0]));
				   ST(&(x[WS(rs, 2)]), VFNMSI(TN, TK), ms, &(x[0]));
				   ST(&(x[WS(rs, 6)]), VFMAI(TP, TO), ms, &(x[0]));
				   ST(&(x[WS(rs, 4)]), VFNMSI(TP, TO), ms, &(x[0]));
				   ST(&(x[WS(rs, 9)]), VFNMSI(Tx, Tu), ms, &(x[WS(rs, 1)]));
				   ST(&(x[WS(rs, 1)]), VFMAI(Tx, Tu), ms, &(x[WS(rs, 1)]));
				   ST(&(x[WS(rs, 7)]), VFNMSI(Tz, Ty), ms, &(x[WS(rs, 1)]));
				   ST(&(x[WS(rs, 3)]), VFMAI(Tz, Ty), ms, &(x[WS(rs, 1)]));
			      }
			 }
		    }
	       }
	  }
     }
     VLEAVE();
}
开发者ID:SKA-ScienceDataProcessor,项目名称:FastImaging,代码行数:98,代码来源:t2bv_10.c

示例8: t1fuv_9

static void t1fuv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP939692620, +0.939692620785908384054109277324731469936208134);
     DVK(KP826351822, +0.826351822333069651148283373230685203999624323);
     DVK(KP879385241, +0.879385241571816768108218554649462939872416269);
     DVK(KP984807753, +0.984807753012208059366743024589523013670643252);
     DVK(KP666666666, +0.666666666666666666666666666666666666666666667);
     DVK(KP852868531, +0.852868531952443209628250963940074071936020296);
     DVK(KP907603734, +0.907603734547952313649323976213898122064543220);
     DVK(KP420276625, +0.420276625461206169731530603237061658838781920);
     DVK(KP673648177, +0.673648177666930348851716626769314796000375677);
     DVK(KP898197570, +0.898197570222573798468955502359086394667167570);
     DVK(KP347296355, +0.347296355333860697703433253538629592000751354);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     DVK(KP439692620, +0.439692620785908384054109277324731469936208134);
     DVK(KP203604859, +0.203604859554852403062088995281827210665664861);
     DVK(KP152703644, +0.152703644666139302296566746461370407999248646);
     DVK(KP586256827, +0.586256827714544512072145703099641959914944179);
     DVK(KP968908795, +0.968908795874236621082202410917456709164223497);
     DVK(KP726681596, +0.726681596905677465811651808188092531873167623);
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     {
	  INT m;
	  R *x;
	  x = ri;
	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
	       V T1, T3, T5, T9, Th, Tb, Td, Tj, Tl, TD, T6;
	       T1 = LD(&(x[0]), ms, &(x[0]));
	       {
		    V T2, T4, T8, Tg;
		    T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
		    T4 = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
		    T8 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
		    Tg = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
		    {
			 V Ta, Tc, Ti, Tk;
			 Ta = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
			 Tc = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
			 Ti = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
			 Tk = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
			 T3 = BYTWJ(&(W[TWVL * 4]), T2);
			 T5 = BYTWJ(&(W[TWVL * 10]), T4);
			 T9 = BYTWJ(&(W[0]), T8);
			 Th = BYTWJ(&(W[TWVL * 2]), Tg);
			 Tb = BYTWJ(&(W[TWVL * 6]), Ta);
			 Td = BYTWJ(&(W[TWVL * 12]), Tc);
			 Tj = BYTWJ(&(W[TWVL * 8]), Ti);
			 Tl = BYTWJ(&(W[TWVL * 14]), Tk);
		    }
	       }
	       TD = VSUB(T5, T3);
	       T6 = VADD(T3, T5);
	       {
		    V Tt, Te, Tu, Tm, Tr, T7;
		    Tt = VSUB(Tb, Td);
		    Te = VADD(Tb, Td);
		    Tu = VSUB(Tl, Tj);
		    Tm = VADD(Tj, Tl);
		    Tr = VFNMS(LDK(KP500000000), T6, T1);
		    T7 = VADD(T1, T6);
		    {
			 V Tv, Tf, Ts, Tn;
			 Tv = VFNMS(LDK(KP500000000), Te, T9);
			 Tf = VADD(T9, Te);
			 Ts = VFNMS(LDK(KP500000000), Tm, Th);
			 Tn = VADD(Th, Tm);
			 {
			      V TG, TK, Tw, TJ, TF, TA, To, Tq;
			      TG = VFNMS(LDK(KP726681596), Tt, Tv);
			      TK = VFMA(LDK(KP968908795), Tv, Tt);
			      Tw = VFNMS(LDK(KP586256827), Tv, Tu);
			      TJ = VFNMS(LDK(KP152703644), Tu, Ts);
			      TF = VFMA(LDK(KP203604859), Ts, Tu);
			      TA = VFNMS(LDK(KP439692620), Tt, Ts);
			      To = VADD(Tf, Tn);
			      Tq = VMUL(LDK(KP866025403), VSUB(Tn, Tf));
			      {
				   V TQ, TH, TL, TN, TB, Tp, Ty, TI, Tx;
				   Tx = VFNMS(LDK(KP347296355), Tw, Tt);
				   TQ = VFNMS(LDK(KP898197570), TG, TF);
				   TH = VFMA(LDK(KP898197570), TG, TF);
				   TL = VFMA(LDK(KP673648177), TK, TJ);
				   TN = VFNMS(LDK(KP673648177), TK, TJ);
				   TB = VFNMS(LDK(KP420276625), TA, Tu);
				   ST(&(x[0]), VADD(T7, To), ms, &(x[0]));
				   Tp = VFNMS(LDK(KP500000000), To, T7);
				   Ty = VFNMS(LDK(KP907603734), Tx, Ts);
				   TI = VFMA(LDK(KP852868531), TH, Tr);
				   {
					V TO, TR, TM, TC, Tz, TP, TS, TE;
					TO = VFNMS(LDK(KP500000000), TH, TN);
					TR = VFMA(LDK(KP666666666), TL, TQ);
					TM = VMUL(LDK(KP984807753), VFNMS(LDK(KP879385241), TD, TL));
					TC = VFNMS(LDK(KP826351822), TB, Tv);
					ST(&(x[WS(rs, 6)]), VFNMSI(Tq, Tp), ms, &(x[0]));
					ST(&(x[WS(rs, 3)]), VFMAI(Tq, Tp), ms, &(x[WS(rs, 1)]));
					Tz = VFNMS(LDK(KP939692620), Ty, Tr);
					TP = VFMA(LDK(KP852868531), TO, Tr);
					TS = VMUL(LDK(KP866025403), VFMA(LDK(KP852868531), TR, TD));
					ST(&(x[WS(rs, 8)]), VFMAI(TM, TI), ms, &(x[0]));
//.........这里部分代码省略.........
开发者ID:dstuck,项目名称:tinker_integrated_PIMC,代码行数:101,代码来源:t1fuv_9.c

示例9: n1fv_7

static void n1fv_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
     DVK(KP900968867, +0.900968867902419126236102319507445051165919162);
     DVK(KP692021471, +0.692021471630095869627814897002069140197260599);
     DVK(KP801937735, +0.801937735804838252472204639014890102331838324);
     DVK(KP974927912, +0.974927912181823607018131682993931217232785801);
     DVK(KP356895867, +0.356895867892209443894399510021300583399127187);
     DVK(KP554958132, +0.554958132087371191422194871006410481067288862);
     INT i;
     const R *xi;
     R *xo;
     xi = ri;
     xo = ro;
     for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
	  V T1, T2, T3, T8, T9, T5, T6;
	  T1 = LD(&(xi[0]), ivs, &(xi[0]));
	  T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
	  T3 = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
	  T8 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
	  T9 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
	  T5 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
	  T6 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
	  {
	       V Te, T4, Tf, Ta, Tg, T7;
	       Te = VSUB(T3, T2);
	       T4 = VADD(T2, T3);
	       Tf = VSUB(T9, T8);
	       Ta = VADD(T8, T9);
	       Tg = VSUB(T6, T5);
	       T7 = VADD(T5, T6);
	       {
		    V Tm, Tb, Tr, Th, Tj, To;
		    Tm = VFMA(LDK(KP554958132), Tf, Te);
		    Tb = VFNMS(LDK(KP356895867), T4, Ta);
		    Tr = VFNMS(LDK(KP554958132), Te, Tg);
		    Th = VFMA(LDK(KP554958132), Tg, Tf);
		    ST(&(xo[0]), VADD(T1, VADD(T4, VADD(T7, Ta))), ovs, &(xo[0]));
		    Tj = VFNMS(LDK(KP356895867), T7, T4);
		    To = VFNMS(LDK(KP356895867), Ta, T7);
		    {
			 V Tn, Tc, Ts, Ti;
			 Tn = VMUL(LDK(KP974927912), VFMA(LDK(KP801937735), Tm, Tg));
			 Tc = VFNMS(LDK(KP692021471), Tb, T7);
			 Ts = VMUL(LDK(KP974927912), VFNMS(LDK(KP801937735), Tr, Tf));
			 Ti = VMUL(LDK(KP974927912), VFNMS(LDK(KP801937735), Th, Te));
			 {
			      V Tk, Tp, Td, Tl, Tq;
			      Tk = VFNMS(LDK(KP692021471), Tj, Ta);
			      Tp = VFNMS(LDK(KP692021471), To, T4);
			      Td = VFNMS(LDK(KP900968867), Tc, T1);
			      Tl = VFNMS(LDK(KP900968867), Tk, T1);
			      Tq = VFNMS(LDK(KP900968867), Tp, T1);
			      ST(&(xo[WS(os, 2)]), VFMAI(Ti, Td), ovs, &(xo[0]));
			      ST(&(xo[WS(os, 5)]), VFNMSI(Ti, Td), ovs, &(xo[WS(os, 1)]));
			      ST(&(xo[WS(os, 1)]), VFMAI(Tn, Tl), ovs, &(xo[WS(os, 1)]));
			      ST(&(xo[WS(os, 6)]), VFNMSI(Tn, Tl), ovs, &(xo[0]));
			      ST(&(xo[WS(os, 3)]), VFMAI(Ts, Tq), ovs, &(xo[WS(os, 1)]));
			      ST(&(xo[WS(os, 4)]), VFNMSI(Ts, Tq), ovs, &(xo[0]));
			 }
		    }
	       }
	  }
     }
}
开发者ID:phillipstanleymarbell,项目名称:sunflower-simulator,代码行数:64,代码来源:n1fv_7.c

示例10: n1bv_12

static void n1bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
    DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
    DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
    {
        INT i;
        const R *xi;
        R *xo;
        xi = ii;
        xo = io;
        for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
            V T5, Ta, TG, TF, Ty, Tm, Ti, Tp, TJ, TI, Tx, Ts;
            {
                V T1, T6, T4, Tk, T9, Tl;
                T1 = LD(&(xi[0]), ivs, &(xi[0]));
                T6 = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
                {
                    V T2, T3, T7, T8;
                    T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
                    T3 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
                    T4 = VADD(T2, T3);
                    Tk = VSUB(T2, T3);
                    T7 = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
                    T8 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
                    T9 = VADD(T7, T8);
                    Tl = VSUB(T7, T8);
                }
                T5 = VFNMS(LDK(KP500000000), T4, T1);
                Ta = VFNMS(LDK(KP500000000), T9, T6);
                TG = VADD(T6, T9);
                TF = VADD(T1, T4);
                Ty = VADD(Tk, Tl);
                Tm = VMUL(LDK(KP866025403), VSUB(Tk, Tl));
            }
            {
                V Tn, Tq, Te, To, Th, Tr;
                Tn = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
                Tq = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
                {
                    V Tc, Td, Tf, Tg;
                    Tc = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
                    Td = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
                    Te = VSUB(Tc, Td);
                    To = VADD(Tc, Td);
                    Tf = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
                    Tg = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
                    Th = VSUB(Tf, Tg);
                    Tr = VADD(Tf, Tg);
                }
                Ti = VMUL(LDK(KP866025403), VSUB(Te, Th));
                Tp = VFNMS(LDK(KP500000000), To, Tn);
                TJ = VADD(Tq, Tr);
                TI = VADD(Tn, To);
                Tx = VADD(Te, Th);
                Ts = VFNMS(LDK(KP500000000), Tr, Tq);
            }
            {
                V TH, TK, TL, TM;
                TH = VSUB(TF, TG);
                TK = VBYI(VSUB(TI, TJ));
                ST(&(xo[WS(os, 3)]), VSUB(TH, TK), ovs, &(xo[WS(os, 1)]));
                ST(&(xo[WS(os, 9)]), VADD(TH, TK), ovs, &(xo[WS(os, 1)]));
                TL = VADD(TF, TG);
                TM = VADD(TI, TJ);
                ST(&(xo[WS(os, 6)]), VSUB(TL, TM), ovs, &(xo[0]));
                ST(&(xo[0]), VADD(TL, TM), ovs, &(xo[0]));
            }
            {
                V Tj, Tv, Tu, Tw, Tb, Tt;
                Tb = VSUB(T5, Ta);
                Tj = VSUB(Tb, Ti);
                Tv = VADD(Tb, Ti);
                Tt = VSUB(Tp, Ts);
                Tu = VBYI(VADD(Tm, Tt));
                Tw = VBYI(VSUB(Tt, Tm));
                ST(&(xo[WS(os, 11)]), VSUB(Tj, Tu), ovs, &(xo[WS(os, 1)]));
                ST(&(xo[WS(os, 5)]), VADD(Tv, Tw), ovs, &(xo[WS(os, 1)]));
                ST(&(xo[WS(os, 1)]), VADD(Tj, Tu), ovs, &(xo[WS(os, 1)]));
                ST(&(xo[WS(os, 7)]), VSUB(Tv, Tw), ovs, &(xo[WS(os, 1)]));
            }
            {
                V Tz, TD, TC, TE, TA, TB;
                Tz = VBYI(VMUL(LDK(KP866025403), VSUB(Tx, Ty)));
                TD = VBYI(VMUL(LDK(KP866025403), VADD(Ty, Tx)));
                TA = VADD(T5, Ta);
                TB = VADD(Tp, Ts);
                TC = VSUB(TA, TB);
                TE = VADD(TA, TB);
                ST(&(xo[WS(os, 2)]), VADD(Tz, TC), ovs, &(xo[0]));
                ST(&(xo[WS(os, 8)]), VSUB(TE, TD), ovs, &(xo[0]));
                ST(&(xo[WS(os, 10)]), VSUB(TC, Tz), ovs, &(xo[0]));
                ST(&(xo[WS(os, 4)]), VADD(TD, TE), ovs, &(xo[0]));
            }
        }
    }
    VLEAVE();
}
开发者ID:bohrasd,项目名称:windowsrtdev,代码行数:97,代码来源:n1bv_12.c

示例11: t2fv_32

static void t2fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP555570233, +0.555570233019602224742830813948532874374937191);
     DVK(KP831469612, +0.831469612302545237078788377617905756738560812);
     DVK(KP195090322, +0.195090322016128267848284868477022240927691618);
     DVK(KP980785280, +0.980785280403230449126182236134239036973933731);
     DVK(KP382683432, +0.382683432365089771728459984030398866761344562);
     DVK(KP923879532, +0.923879532511286756128183189396788286822416626);
     DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
     {
	  INT m;
	  R *x;
	  x = ri;
	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
	       V T4, T1A, T2o, T32, Tf, T1v, T2r, T3f, TC, T1C, T2L, T34, Tr, T1D, T2O;
	       V T33, T1k, T20, T2F, T3b, T1r, T21, T2C, T3a, TV, T1X, T2y, T38, T12, T1Y;
	       V T2v, T37;
	       {
		    V T1, T1z, T3, T1x, T1y, T2, T1w, T2m, T2n;
		    T1 = LD(&(x[0]), ms, &(x[0]));
		    T1y = LD(&(x[WS(rs, 24)]), ms, &(x[0]));
		    T1z = BYTWJ(&(W[TWVL * 46]), T1y);
		    T2 = LD(&(x[WS(rs, 16)]), ms, &(x[0]));
		    T3 = BYTWJ(&(W[TWVL * 30]), T2);
		    T1w = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
		    T1x = BYTWJ(&(W[TWVL * 14]), T1w);
		    T4 = VSUB(T1, T3);
		    T1A = VSUB(T1x, T1z);
		    T2m = VADD(T1, T3);
		    T2n = VADD(T1x, T1z);
		    T2o = VADD(T2m, T2n);
		    T32 = VSUB(T2m, T2n);
	       }
	       {
		    V T6, Td, T8, Tb;
		    {
			 V T5, Tc, T7, Ta;
			 T5 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
			 T6 = BYTWJ(&(W[TWVL * 6]), T5);
			 Tc = LD(&(x[WS(rs, 12)]), ms, &(x[0]));
			 Td = BYTWJ(&(W[TWVL * 22]), Tc);
			 T7 = LD(&(x[WS(rs, 20)]), ms, &(x[0]));
			 T8 = BYTWJ(&(W[TWVL * 38]), T7);
			 Ta = LD(&(x[WS(rs, 28)]), ms, &(x[0]));
			 Tb = BYTWJ(&(W[TWVL * 54]), Ta);
		    }
		    {
			 V T9, Te, T2p, T2q;
			 T9 = VSUB(T6, T8);
			 Te = VSUB(Tb, Td);
			 Tf = VMUL(LDK(KP707106781), VADD(T9, Te));
			 T1v = VMUL(LDK(KP707106781), VSUB(Te, T9));
			 T2p = VADD(T6, T8);
			 T2q = VADD(Tb, Td);
			 T2r = VADD(T2p, T2q);
			 T3f = VSUB(T2q, T2p);
		    }
	       }
	       {
		    V Tt, TA, Tv, Ty;
		    {
			 V Ts, Tz, Tu, Tx;
			 Ts = LD(&(x[WS(rs, 30)]), ms, &(x[0]));
			 Tt = BYTWJ(&(W[TWVL * 58]), Ts);
			 Tz = LD(&(x[WS(rs, 22)]), ms, &(x[0]));
			 TA = BYTWJ(&(W[TWVL * 42]), Tz);
			 Tu = LD(&(x[WS(rs, 14)]), ms, &(x[0]));
			 Tv = BYTWJ(&(W[TWVL * 26]), Tu);
			 Tx = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
			 Ty = BYTWJ(&(W[TWVL * 10]), Tx);
		    }
		    {
			 V Tw, TB, T2J, T2K;
			 Tw = VSUB(Tt, Tv);
			 TB = VSUB(Ty, TA);
			 TC = VFMA(LDK(KP923879532), Tw, VMUL(LDK(KP382683432), TB));
			 T1C = VFNMS(LDK(KP923879532), TB, VMUL(LDK(KP382683432), Tw));
			 T2J = VADD(Tt, Tv);
			 T2K = VADD(Ty, TA);
			 T2L = VADD(T2J, T2K);
			 T34 = VSUB(T2J, T2K);
		    }
	       }
	       {
		    V Ti, Tp, Tk, Tn;
		    {
			 V Th, To, Tj, Tm;
			 Th = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
			 Ti = BYTWJ(&(W[TWVL * 2]), Th);
			 To = LD(&(x[WS(rs, 26)]), ms, &(x[0]));
			 Tp = BYTWJ(&(W[TWVL * 50]), To);
			 Tj = LD(&(x[WS(rs, 18)]), ms, &(x[0]));
			 Tk = BYTWJ(&(W[TWVL * 34]), Tj);
			 Tm = LD(&(x[WS(rs, 10)]), ms, &(x[0]));
			 Tn = BYTWJ(&(W[TWVL * 18]), Tm);
		    }
		    {
			 V Tl, Tq, T2M, T2N;
			 Tl = VSUB(Ti, Tk);
			 Tq = VSUB(Tn, Tp);
//.........这里部分代码省略.........
开发者ID:AlexanderB1290,项目名称:scafacos,代码行数:101,代码来源:t2fv_32.c

示例12: hc2cfdftv_8

static void hc2cfdftv_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
     {
	  INT m;
	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(32, rs)) {
	       V T3, Tc, Tl, Ts, Tf, Tg, Te, Tp, T7, Ta, T1, T2, Tb, Tj, Tk;
	       V Ti, Tr, T5, T6, T4, T9, Th, Tq, TC, T8, Td, TF, Tm, TG, TD;
	       V Tt, Tu, Tn, TH, TL, TE, TK, Tz, Tv, Ty, To, TJ, TI, TN, TM;
	       V TB, TA, Tx, Tw;
	       T1 = LD(&(Rp[0]), ms, &(Rp[0]));
	       T2 = LD(&(Rm[0]), -ms, &(Rm[0]));
	       Tb = LDW(&(W[0]));
	       Tj = LD(&(Rp[WS(rs, 3)]), ms, &(Rp[WS(rs, 1)]));
	       Tk = LD(&(Rm[WS(rs, 3)]), -ms, &(Rm[WS(rs, 1)]));
	       Ti = LDW(&(W[TWVL * 12]));
	       Tr = LDW(&(W[TWVL * 10]));
	       T5 = LD(&(Rp[WS(rs, 2)]), ms, &(Rp[0]));
	       T6 = LD(&(Rm[WS(rs, 2)]), -ms, &(Rm[0]));
	       T3 = VFMACONJ(T2, T1);
	       Tc = VZMULIJ(Tb, VFNMSCONJ(T2, T1));
	       T4 = LDW(&(W[TWVL * 6]));
	       T9 = LDW(&(W[TWVL * 8]));
	       Tl = VZMULIJ(Ti, VFNMSCONJ(Tk, Tj));
	       Ts = VZMULJ(Tr, VFMACONJ(Tk, Tj));
	       Tf = LD(&(Rp[WS(rs, 1)]), ms, &(Rp[WS(rs, 1)]));
	       Tg = LD(&(Rm[WS(rs, 1)]), -ms, &(Rm[WS(rs, 1)]));
	       Te = LDW(&(W[TWVL * 4]));
	       Tp = LDW(&(W[TWVL * 2]));
	       T7 = VZMULJ(T4, VFMACONJ(T6, T5));
	       Ta = VZMULIJ(T9, VFNMSCONJ(T6, T5));
	       Th = VZMULIJ(Te, VFNMSCONJ(Tg, Tf));
	       Tq = VZMULJ(Tp, VFMACONJ(Tg, Tf));
	       TC = VADD(T3, T7);
	       T8 = VSUB(T3, T7);
	       Td = VSUB(Ta, Tc);
	       TF = VADD(Tc, Ta);
	       Tm = VSUB(Th, Tl);
	       TG = VADD(Th, Tl);
	       TD = VADD(Tq, Ts);
	       Tt = VSUB(Tq, Ts);
	       Tu = VSUB(Tm, Td);
	       Tn = VADD(Td, Tm);
	       TH = VSUB(TF, TG);
	       TL = VADD(TF, TG);
	       TE = VSUB(TC, TD);
	       TK = VADD(TC, TD);
	       Tz = VFMA(LDK(KP707106781), Tu, Tt);
	       Tv = VFNMS(LDK(KP707106781), Tu, Tt);
	       Ty = VFNMS(LDK(KP707106781), Tn, T8);
	       To = VFMA(LDK(KP707106781), Tn, T8);
	       TJ = VCONJ(VMUL(LDK(KP500000000), VFNMSI(TH, TE)));
	       TI = VMUL(LDK(KP500000000), VFMAI(TH, TE));
	       TN = VCONJ(VMUL(LDK(KP500000000), VADD(TL, TK)));
	       TM = VMUL(LDK(KP500000000), VSUB(TK, TL));
	       TB = VMUL(LDK(KP500000000), VFMAI(Tz, Ty));
	       TA = VCONJ(VMUL(LDK(KP500000000), VFNMSI(Tz, Ty)));
	       Tx = VCONJ(VMUL(LDK(KP500000000), VFMAI(Tv, To)));
	       Tw = VMUL(LDK(KP500000000), VFNMSI(Tv, To));
	       ST(&(Rm[WS(rs, 1)]), TJ, -ms, &(Rm[WS(rs, 1)]));
	       ST(&(Rp[WS(rs, 2)]), TI, ms, &(Rp[0]));
	       ST(&(Rm[WS(rs, 3)]), TN, -ms, &(Rm[WS(rs, 1)]));
	       ST(&(Rp[0]), TM, ms, &(Rp[0]));
	       ST(&(Rp[WS(rs, 3)]), TB, ms, &(Rp[WS(rs, 1)]));
	       ST(&(Rm[WS(rs, 2)]), TA, -ms, &(Rm[0]));
	       ST(&(Rm[0]), Tx, -ms, &(Rm[0]));
	       ST(&(Rp[WS(rs, 1)]), Tw, ms, &(Rp[WS(rs, 1)]));
	  }
     }
     VLEAVE();
}
开发者ID:barnex,项目名称:fftw,代码行数:72,代码来源:rdft_simd_sse2_hc2cfdftv_8.c

示例13: n2sv_16


//.........这里部分代码省略.........
		    STM4(&(ro[8]), T2l, ovs, &(ro[0]));
		    T2m = VADD(Tf, Tu);
		    STM4(&(ro[0]), T2m, ovs, &(ro[0]));
		    T2j = VADD(TC, TJ);
		    T2k = VADD(T2g, T2h);
		    T2n = VSUB(T2j, T2k);
		    STM4(&(io[8]), T2n, ovs, &(io[0]));
		    T2o = VADD(T2j, T2k);
		    STM4(&(io[0]), T2o, ovs, &(io[0]));
	       }
	       {
		    V Tv, TK, T2f, T2i;
		    Tv = VSUB(Tt, Tm);
		    TK = VSUB(TC, TJ);
		    T2p = VADD(Tv, TK);
		    STM4(&(io[4]), T2p, ovs, &(io[0]));
		    T2q = VSUB(TK, Tv);
		    STM4(&(io[12]), T2q, ovs, &(io[0]));
		    T2f = VSUB(T7, Te);
		    T2i = VSUB(T2g, T2h);
		    T2r = VSUB(T2f, T2i);
		    STM4(&(ro[12]), T2r, ovs, &(ro[0]));
		    T2s = VADD(T2f, T2i);
		    STM4(&(ro[4]), T2s, ovs, &(ro[0]));
	       }
	       {
		    V T2t, T2u, T2v, T2w, T2x, T2y, T2z, T2A;
		    {
			 V T1T, T27, T24, T28, T1Y, T23;
			 T1T = VADD(T1R, T1S);
			 T27 = VSUB(T25, T26);
			 T1Y = VADD(T1U, T1X);
			 T23 = VSUB(T1Z, T22);
			 T24 = VMUL(LDK(KP707106781), VADD(T1Y, T23));
			 T28 = VMUL(LDK(KP707106781), VSUB(T23, T1Y));
			 T2t = VSUB(T1T, T24);
			 STM4(&(ro[10]), T2t, ovs, &(ro[0]));
			 T2u = VADD(T27, T28);
			 STM4(&(io[6]), T2u, ovs, &(io[0]));
			 T2v = VADD(T1T, T24);
			 STM4(&(ro[2]), T2v, ovs, &(ro[0]));
			 T2w = VSUB(T27, T28);
			 STM4(&(io[14]), T2w, ovs, &(io[0]));
		    }
		    {
			 V T29, T2d, T2c, T2e, T2a, T2b;
			 T29 = VSUB(T1R, T1S);
			 T2d = VADD(T26, T25);
			 T2a = VSUB(T1X, T1U);
			 T2b = VADD(T1Z, T22);
			 T2c = VMUL(LDK(KP707106781), VSUB(T2a, T2b));
			 T2e = VMUL(LDK(KP707106781), VADD(T2a, T2b));
			 T2x = VSUB(T29, T2c);
			 STM4(&(ro[14]), T2x, ovs, &(ro[0]));
			 T2y = VADD(T2d, T2e);
			 STM4(&(io[2]), T2y, ovs, &(io[0]));
			 T2z = VADD(T29, T2c);
			 STM4(&(ro[6]), T2z, ovs, &(ro[0]));
			 T2A = VSUB(T2d, T2e);
			 STM4(&(io[10]), T2A, ovs, &(io[0]));
		    }
		    {
			 V T2B, T2C, T2D, T2E, T2F, T2G, T2H, T2I;
			 {
			      V TV, T1r, T1p, T1v, T1i, T1q, T1u, T1w, TU, T1o;
			      TU = VMUL(LDK(KP707106781), VSUB(TQ, TT));
开发者ID:8cH9azbsFifZ,项目名称:wspr,代码行数:67,代码来源:n2sv_16.c

示例14: n2bv_10

static void n2bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     INT i;
     const R *xi;
     R *xo;
     xi = ii;
     xo = io;
     for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
	  V Tl, Ty, T7, Te, Tw, Tt, Tz, TA, TB, Tg, Th, Tm, Tj, Tk;
	  Tj = LD(&(xi[0]), ivs, &(xi[0]));
	  Tk = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
	  Tl = VSUB(Tj, Tk);
	  Ty = VADD(Tj, Tk);
	  {
	       V T3, Tr, Td, Tv, T6, Ts, Ta, Tu;
	       {
		    V T1, T2, Tb, Tc;
		    T1 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
		    T2 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
		    T3 = VSUB(T1, T2);
		    Tr = VADD(T1, T2);
		    Tb = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
		    Tc = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
		    Td = VSUB(Tb, Tc);
		    Tv = VADD(Tb, Tc);
	       }
	       {
		    V T4, T5, T8, T9;
		    T4 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
		    T5 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
		    T6 = VSUB(T4, T5);
		    Ts = VADD(T4, T5);
		    T8 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
		    T9 = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
		    Ta = VSUB(T8, T9);
		    Tu = VADD(T8, T9);
	       }
	       T7 = VSUB(T3, T6);
	       Te = VSUB(Ta, Td);
	       Tw = VSUB(Tu, Tv);
	       Tt = VSUB(Tr, Ts);
	       Tz = VADD(Tr, Ts);
	       TA = VADD(Tu, Tv);
	       TB = VADD(Tz, TA);
	       Tg = VADD(T3, T6);
	       Th = VADD(Ta, Td);
	       Tm = VADD(Tg, Th);
	  }
	  {
	       V TH, TI, TK, TL, TM;
	       TH = VADD(Tl, Tm);
	       STM2(&(xo[10]), TH, ovs, &(xo[2]));
	       TI = VADD(Ty, TB);
	       STM2(&(xo[0]), TI, ovs, &(xo[0]));
	       {
		    V Tf, Tq, To, Tp, Ti, Tn, TJ;
		    Tf = VBYI(VFMA(LDK(KP951056516), T7, VMUL(LDK(KP587785252), Te)));
		    Tq = VBYI(VFNMS(LDK(KP951056516), Te, VMUL(LDK(KP587785252), T7)));
		    Ti = VMUL(LDK(KP559016994), VSUB(Tg, Th));
		    Tn = VFNMS(LDK(KP250000000), Tm, Tl);
		    To = VADD(Ti, Tn);
		    Tp = VSUB(Tn, Ti);
		    TJ = VADD(Tf, To);
		    STM2(&(xo[2]), TJ, ovs, &(xo[2]));
		    STN2(&(xo[0]), TI, TJ, ovs);
		    TK = VADD(Tq, Tp);
		    STM2(&(xo[14]), TK, ovs, &(xo[2]));
		    TL = VSUB(To, Tf);
		    STM2(&(xo[18]), TL, ovs, &(xo[2]));
		    TM = VSUB(Tp, Tq);
		    STM2(&(xo[6]), TM, ovs, &(xo[2]));
	       }
	       {
		    V Tx, TG, TE, TF, TC, TD;
		    Tx = VBYI(VFNMS(LDK(KP951056516), Tw, VMUL(LDK(KP587785252), Tt)));
		    TG = VBYI(VFMA(LDK(KP951056516), Tt, VMUL(LDK(KP587785252), Tw)));
		    TC = VFNMS(LDK(KP250000000), TB, Ty);
		    TD = VMUL(LDK(KP559016994), VSUB(Tz, TA));
		    TE = VSUB(TC, TD);
		    TF = VADD(TD, TC);
		    {
			 V TN, TO, TP, TQ;
			 TN = VADD(Tx, TE);
			 STM2(&(xo[4]), TN, ovs, &(xo[0]));
			 STN2(&(xo[4]), TN, TM, ovs);
			 TO = VADD(TG, TF);
			 STM2(&(xo[12]), TO, ovs, &(xo[0]));
			 STN2(&(xo[12]), TO, TK, ovs);
			 TP = VSUB(TE, Tx);
			 STM2(&(xo[16]), TP, ovs, &(xo[0]));
			 STN2(&(xo[16]), TP, TL, ovs);
			 TQ = VSUB(TF, TG);
			 STM2(&(xo[8]), TQ, ovs, &(xo[0]));
			 STN2(&(xo[8]), TQ, TH, ovs);
		    }
	       }
//.........这里部分代码省略.........
开发者ID:8cH9azbsFifZ,项目名称:wspr,代码行数:101,代码来源:n2bv_10.c

示例15: n1fv_13


//.........这里部分代码省略.........
		    TK = VADD(Tx, Ty);
		    Tz = VSUB(Tx, Ty);
		    Tc = VADD(T6, Tb);
		    TX = VSUB(T6, Tb);
		    Tv = VSUB(Tt, Tu);
		    TO = VADD(Tt, Tu);
		    TL = VSUB(Ti, Tl);
		    Tm = VADD(Ti, Tl);
		    {
			 V TF, Tw, TP, TY, TT, TM, TA, Tn;
			 TF = VSUB(Ts, Tv);
			 Tw = VADD(Ts, Tv);
			 TP = VFNMS(LDK(KP500000000), TO, TN);
			 TY = VADD(TN, TO);
			 TT = VFNMS(LDK(KP866025403), TL, TK);
			 TM = VFMA(LDK(KP866025403), TL, TK);
			 TA = VFNMS(LDK(KP500000000), Tm, Tf);
			 Tn = VADD(Tf, Tm);
			 {
			      V T1f, T1n, TI, T18, T1k, T1c, TD, T17, T10, T1m, T16, T1e, TU, TR;
			      TU = VFNMS(LDK(KP866025403), TQ, TP);
			      TR = VFMA(LDK(KP866025403), TQ, TP);
			      {
				   V TZ, T15, TE, TB;
				   TZ = VFMA(LDK(KP302775637), TY, TX);
				   T15 = VFNMS(LDK(KP302775637), TX, TY);
				   TE = VSUB(Tz, TA);
				   TB = VADD(Tz, TA);
				   {
					V TH, To, TV, T13;
					TH = VSUB(Tc, Tn);
					To = VADD(Tc, Tn);
					TV = VFNMS(LDK(KP612264650), TU, TT);
					T13 = VFMA(LDK(KP612264650), TT, TU);
					{
					     V TS, T12, TG, T1b;
					     TS = VFNMS(LDK(KP038632954), TR, TM);
					     T12 = VFMA(LDK(KP038632954), TM, TR);
					     TG = VFNMS(LDK(KP514918778), TF, TE);
					     T1b = VFMA(LDK(KP686558370), TE, TF);
					     {
						  V TC, T1a, Tp, TW, T14;
						  TC = VFMA(LDK(KP301479260), TB, Tw);
						  T1a = VFNMS(LDK(KP226109445), Tw, TB);
						  Tp = VFNMS(LDK(KP083333333), To, T1);
						  ST(&(xo[0]), VADD(T1, To), ovs, &(xo[0]));
						  T1f = VFMA(LDK(KP853480001), TV, TS);
						  TW = VFNMS(LDK(KP853480001), TV, TS);
						  T1n = VFMA(LDK(KP853480001), T13, T12);
						  T14 = VFNMS(LDK(KP853480001), T13, T12);
						  TI = VFMA(LDK(KP581704778), TH, TG);
						  T18 = VFNMS(LDK(KP859542535), TG, TH);
						  T1k = VFMA(LDK(KP769338817), T1b, T1a);
						  T1c = VFNMS(LDK(KP769338817), T1b, T1a);
						  TD = VFMA(LDK(KP503537032), TC, Tp);
						  T17 = VFNMS(LDK(KP251768516), TC, Tp);
						  T10 = VMUL(LDK(KP600477271), VFMA(LDK(KP957805992), TZ, TW));
						  T1m = VFNMS(LDK(KP522026385), TW, TZ);
						  T16 = VMUL(LDK(KP600477271), VFMA(LDK(KP957805992), T15, T14));
						  T1e = VFNMS(LDK(KP522026385), T14, T15);
					     }
					}
				   }
			      }
			      {
				   V T1o, T1q, T1g, T1i, T1d, T1h, T1l, T1p;
				   {
					V T11, TJ, T19, T1j;
					T11 = VFMA(LDK(KP516520780), TI, TD);
					TJ = VFNMS(LDK(KP516520780), TI, TD);
					T19 = VFMA(LDK(KP300462606), T18, T17);
					T1j = VFNMS(LDK(KP300462606), T18, T17);
					T1o = VMUL(LDK(KP575140729), VFNMS(LDK(KP904176221), T1n, T1m));
					T1q = VMUL(LDK(KP575140729), VFMA(LDK(KP904176221), T1n, T1m));
					T1g = VMUL(LDK(KP575140729), VFMA(LDK(KP904176221), T1f, T1e));
					T1i = VMUL(LDK(KP575140729), VFNMS(LDK(KP904176221), T1f, T1e));
					ST(&(xo[WS(os, 12)]), VFNMSI(T16, T11), ovs, &(xo[0]));
					ST(&(xo[WS(os, 1)]), VFMAI(T16, T11), ovs, &(xo[WS(os, 1)]));
					ST(&(xo[WS(os, 8)]), VFMAI(T10, TJ), ovs, &(xo[0]));
					ST(&(xo[WS(os, 5)]), VFNMSI(T10, TJ), ovs, &(xo[WS(os, 1)]));
					T1d = VFNMS(LDK(KP503537032), T1c, T19);
					T1h = VFMA(LDK(KP503537032), T1c, T19);
					T1l = VFNMS(LDK(KP503537032), T1k, T1j);
					T1p = VFMA(LDK(KP503537032), T1k, T1j);
				   }
				   ST(&(xo[WS(os, 9)]), VFMAI(T1g, T1d), ovs, &(xo[WS(os, 1)]));
				   ST(&(xo[WS(os, 4)]), VFNMSI(T1g, T1d), ovs, &(xo[0]));
				   ST(&(xo[WS(os, 10)]), VFNMSI(T1i, T1h), ovs, &(xo[0]));
				   ST(&(xo[WS(os, 3)]), VFMAI(T1i, T1h), ovs, &(xo[WS(os, 1)]));
				   ST(&(xo[WS(os, 7)]), VFMAI(T1o, T1l), ovs, &(xo[WS(os, 1)]));
				   ST(&(xo[WS(os, 6)]), VFNMSI(T1o, T1l), ovs, &(xo[0]));
				   ST(&(xo[WS(os, 11)]), VFMAI(T1q, T1p), ovs, &(xo[WS(os, 1)]));
				   ST(&(xo[WS(os, 2)]), VFNMSI(T1q, T1p), ovs, &(xo[0]));
			      }
			 }
		    }
	       }
	  }
     }
}
开发者ID:bambang,项目名称:vsipl,代码行数:101,代码来源:n1fv_13.c


注:本文中的VMUL函数示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。