Exemplo n.º 1
0
// merge two bodies
body body_merge(body a, body b) {
  body merged;
  merged.m = a.m + b.m;
  merged.pos = vec_add(a.pos, b.pos);
  merged.vel = vec_div(vec_add(vec_mul(a.pos, a.m), vec_mul(a.pos, b.m)), b.m);

  return merged;
}
Exemplo n.º 2
0
__SIMDd _SIMD_div_pd(__SIMDd a, __SIMDd b)
{
#ifdef  USE_SSE
  return _mm_div_pd(a,b);
#elif defined USE_AVX
  return _m256_div_ps(a,b);
#elif defined USE_IBM
  return vec_div(a,b);
#endif
}
Exemplo n.º 3
0
Arquivo: model.c Projeto: ccxvii/mio
struct model *load_model(const char *name)
{
	char filename[1024];
	unsigned char *data = NULL;
	struct model *model;
	int i, len;

	model = lookup(model_cache, name);
	if (model)
		return model;

	for (i = 0; i < nelem(formats); i++) {
		strlcpy(filename, name, sizeof filename);
		strlcat(filename, formats[i].suffix, sizeof filename);
		data = load_file(filename, &len);
		if (data)
			break;
	}

	if (!data) {
		warn("error: cannot find model: '%s'", name);
		return NULL;
	}

	model = formats[i].load(filename, data, len);
	if (!model)
		warn("error: cannot load model: '%s'", filename);

	free(data);

	if (model)
		model_cache = insert(model_cache, name, model);

	if (model && model->anim) {
		struct anim *anim = model->anim;
		struct pose a_from_org, b_from_org;
		vec4 org_from_a;
		extract_frame_root(&a_from_org, anim, 0);
		extract_frame_root(&b_from_org, anim, anim->frames - 1);
		vec_sub(anim->motion.position, b_from_org.position, a_from_org.position);
		quat_conjugate(org_from_a, a_from_org.rotation);
		quat_mul(anim->motion.rotation, b_from_org.rotation, org_from_a);
		vec_div(anim->motion.scale, b_from_org.scale, a_from_org.scale);
	}

	return model;
}
Exemplo n.º 4
0
int main ()
{
  vector float fa = {1.0, 2.0, 3.0, -4.0};
  vector float fb = {-2.0, -3.0, -4.0, -5.0};
  vector float fc = vec_cpsgn (fa, fb);

  vector long long la = {5L, 14L};
  vector long long lb = {3L, 86L};
  vector long long lc = vec_and (la, lb);
  vector bool long long ld = {0, -1};
  vector long long le = vec_and (la, ld);
  vector long long lf = vec_and (ld, lb);

  vector unsigned long long ua = {5L, 14L};
  vector unsigned long long ub = {3L, 86L};
  vector unsigned long long uc = vec_and (ua, ub);
  vector bool long long ud = {0, -1};
  vector unsigned long long ue = vec_and (ua, ud);
  vector unsigned long long uf = vec_and (ud, ub);

  vector long long lg = vec_andc (la, lb);
  vector long long lh = vec_andc (la, ld);
  vector long long li = vec_andc (ld, lb);

  vector unsigned long long ug = vec_andc (ua, ub);
  vector unsigned long long uh = vec_andc (ua, ud);
  vector unsigned long long ui = vec_andc (ud, ub);

  vector double da = {1.0, -4.0};
  vector double db = {-2.0, 5.0};
  vector double dc = vec_cpsgn (da, db);

  vector long long lj = vec_mergeh (la, lb);
  vector long long lk = vec_mergeh (la, ld);
  vector long long ll = vec_mergeh (ld, la);

  vector unsigned long long uj = vec_mergeh (ua, ub);
  vector unsigned long long uk = vec_mergeh (ua, ud);
  vector unsigned long long ul = vec_mergeh (ud, ua);

  vector long long lm = vec_mergel (la, lb);
  vector long long ln = vec_mergel (la, ld);
  vector long long lo = vec_mergel (ld, la);

  vector unsigned long long um = vec_mergel (ua, ub);
  vector unsigned long long un = vec_mergel (ua, ud);
  vector unsigned long long uo = vec_mergel (ud, ua);

  vector long long lp = vec_nor (la, lb);
  vector long long lq = vec_nor (la, ld);
  vector long long lr = vec_nor (ld, la);

  vector unsigned long long up = vec_nor (ua, ub);
  vector unsigned long long uq = vec_nor (ua, ud);
  vector unsigned long long ur = vec_nor (ud, ua);

  vector long long ls = vec_or (la, lb);
  vector long long lt = vec_or (la, ld);
  vector long long lu = vec_or (ld, la);

  vector unsigned long long us = vec_or (ua, ub);
  vector unsigned long long ut = vec_or (ua, ud);
  vector unsigned long long uu = vec_or (ud, ua);

  vector unsigned char ca = {0,4,8,1,5,9,2,6,10,3,7,11,15,12,14,13};
  vector long long lv = vec_perm (la, lb, ca);
  vector unsigned long long uv = vec_perm (ua, ub, ca);

  vector long long lw = vec_sel (la, lb, lc);
  vector long long lx = vec_sel (la, lb, uc);
  vector long long ly = vec_sel (la, lb, ld);

  vector unsigned long long uw = vec_sel (ua, ub, lc);
  vector unsigned long long ux = vec_sel (ua, ub, uc);
  vector unsigned long long uy = vec_sel (ua, ub, ld);

  vector long long lz = vec_xor (la, lb);
  vector long long l0 = vec_xor (la, ld);
  vector long long l1 = vec_xor (ld, la);

  vector unsigned long long uz = vec_xor (ua, ub);
  vector unsigned long long u0 = vec_xor (ua, ud);
  vector unsigned long long u1 = vec_xor (ud, ua);

  int ia = vec_all_eq (ua, ub);
  int ib = vec_all_ge (ua, ub);
  int ic = vec_all_gt (ua, ub);
  int id = vec_all_le (ua, ub);
  int ie = vec_all_lt (ua, ub);
  int ig = vec_all_ne (ua, ub);

  int ih = vec_any_eq (ua, ub);
  int ii = vec_any_ge (ua, ub);
  int ij = vec_any_gt (ua, ub);
  int ik = vec_any_le (ua, ub);
  int il = vec_any_lt (ua, ub);
  int im = vec_any_ne (ua, ub);

  vector int sia = {9, 16, 25, 36};
  vector int sib = {-8, -27, -64, -125};
  vector int sic = vec_mergee (sia, sib);
  vector int sid = vec_mergeo (sia, sib);

  vector unsigned int uia = {9, 16, 25, 36};
  vector unsigned int uib = {8, 27, 64, 125};
  vector unsigned int uic = vec_mergee (uia, uib);
  vector unsigned int uid = vec_mergeo (uia, uib);

  vector bool int bia = {0, -1, -1, 0};
  vector bool int bib = {-1, -1, 0, -1};
  vector bool int bic = vec_mergee (bia, bib);
  vector bool int bid = vec_mergeo (bia, bib);

  vector unsigned int uie = vec_packsu (ua, ub);

  vector long long l2 = vec_cntlz (la);
  vector unsigned long long u2 = vec_cntlz (ua);
  vector int sie = vec_cntlz (sia);
  vector unsigned int uif = vec_cntlz (uia);
  vector short ssa = {20, -40, -60, 80, 100, -120, -140, 160};
  vector short ssb = vec_cntlz (ssa);
  vector unsigned short usa = {81, 72, 63, 54, 45, 36, 27, 18};
  vector unsigned short usb = vec_cntlz (usa);
  vector signed char sca = {-4, 3, -9, 15, -31, 31, 0, 0,
		            1, 117, -36, 99, 98, 97, 96, 95};
  vector signed char scb = vec_cntlz (sca);
  vector unsigned char cb = vec_cntlz (ca);

  vector double dd = vec_xl (0, &y);
  vec_xst (dd, 0, &z);

  vector double de = vec_round (dd);

  vector double df = vec_splat (de, 0);
  vector double dg = vec_splat (de, 1);
  vector long long l3 = vec_splat (l2, 0);
  vector long long l4 = vec_splat (l2, 1);
  vector unsigned long long u3 = vec_splat (u2, 0);
  vector unsigned long long u4 = vec_splat (u2, 1);
  vector bool long long l5 = vec_splat (ld, 0);
  vector bool long long l6 = vec_splat (ld, 1);

  vector long long l7 = vec_div (l3, l4);
  vector unsigned long long u5 = vec_div (u3, u4);

  vector long long l8 = vec_mul (l3, l4);
  vector unsigned long long u6 = vec_mul (u3, u4);

  vector double dh = vec_ctf (la, -2);
  vector double di = vec_ctf (ua, 2);
  vector long long l9 = vec_cts (dh, -2);
  vector unsigned long long u7 = vec_ctu (di, 2);

  return 0;
}
Exemplo n.º 5
0
void test1() {
// CHECK-LABEL: define void @test1
// CHECK-LE-LABEL: define void @test1

  res_vf = vec_abs(vf);
// CHECK: call <4 x float> @llvm.fabs.v4f32(<4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x float> @llvm.fabs.v4f32(<4 x float> %{{[0-9]*}})

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_add(vd, vd);
// CHECK: fadd <2 x double>
// CHECK-LE: fadd <2 x double>

  res_vd = vec_and(vbll, vd);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>
// CHECK-LE: and <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  res_vd = vec_and(vd, vbll);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>
// CHECK-LE: and <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  res_vd = vec_and(vd, vd);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>
// CHECK-LE: and <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_andc(vbll, vd);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>
// CHECK-LE: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK-LE: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK-LE: and <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_andc(vd, vbll);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>
// CHECK-LE: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK-LE: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK-LE: and <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_andc(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_ceil(vd);
// CHECK: call <2 x double> @llvm.ceil.v2f64(<2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x double> @llvm.ceil.v2f64(<2 x double> %{{[0-9]*}})

  res_vf = vec_ceil(vf);
// CHECK: call <4 x float> @llvm.ceil.v4f32(<4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x float> @llvm.ceil.v4f32(<4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpeq(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpeqdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x i64> @llvm.ppc.vsx.xvcmpeqdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpeq(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpeqsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x i32> @llvm.ppc.vsx.xvcmpeqsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpge(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpge(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpgt(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpgt(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmple(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmple(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmplt(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})
// CHECK-LE: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmplt(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})
// CHECK-LE: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  /* vec_cpsgn */
  res_vf = vec_cpsgn(vf, vf);
// CHECK: call <4 x float> @llvm.copysign.v4f32(<4 x float> %{{.+}}, <4 x float> %{{.+}})
// CHECK-LE: call <4 x float> @llvm.copysign.v4f32(<4 x float> %{{.+}}, <4 x float> %{{.+}})

  res_vd = vec_cpsgn(vd, vd);
// CHECK: call <2 x double> @llvm.copysign.v2f64(<2 x double> %{{.+}}, <2 x double> %{{.+}})
// CHECK-LE: call <2 x double> @llvm.copysign.v2f64(<2 x double> %{{.+}}, <2 x double> %{{.+}})

  /* vec_div */
  res_vsll = vec_div(vsll, vsll);
// CHECK: sdiv <2 x i64>
// CHECK-LE: sdiv <2 x i64>

  res_vull = vec_div(vull, vull);
// CHECK: udiv <2 x i64>
// CHECK-LE: udiv <2 x i64>

  res_vf = vec_div(vf, vf);
// CHECK: fdiv <4 x float>
// CHECK-LE: fdiv <4 x float>

  res_vd = vec_div(vd, vd);
// CHECK: fdiv <2 x double>
// CHECK-LE: fdiv <2 x double>

  /* vec_max */
  res_vf = vec_max(vf, vf);
// CHECK: @llvm.ppc.vsx.xvmaxsp
// CHECK-LE: @llvm.ppc.vsx.xvmaxsp

  res_vd = vec_max(vd, vd);
// CHECK: @llvm.ppc.vsx.xvmaxdp
// CHECK-LE: @llvm.ppc.vsx.xvmaxdp

  res_vf = vec_vmaxfp(vf, vf);
// CHECK: @llvm.ppc.vsx.xvmaxsp
// CHECK-LE: @llvm.ppc.vsx.xvmaxsp

  /* vec_min */
  res_vf = vec_min(vf, vf);
// CHECK: @llvm.ppc.vsx.xvminsp
// CHECK-LE: @llvm.ppc.vsx.xvminsp

  res_vd = vec_min(vd, vd);
// CHECK: @llvm.ppc.vsx.xvmindp
// CHECK-LE: @llvm.ppc.vsx.xvmindp

  res_vf = vec_vminfp(vf, vf);
// CHECK: @llvm.ppc.vsx.xvminsp
// CHECK-LE: @llvm.ppc.vsx.xvminsp

  res_d = __builtin_vsx_xsmaxdp(d, d);
// CHECK: @llvm.ppc.vsx.xsmaxdp
// CHECK-LE: @llvm.ppc.vsx.xsmaxdp

  res_d = __builtin_vsx_xsmindp(d, d);
// CHECK: @llvm.ppc.vsx.xsmindp
// CHECK-LE: @llvm.ppc.vsx.xsmindp

  /* vec_perm */
  res_vsll = vec_perm(vsll, vsll, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_perm(vull, vull, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vbll = vec_perm(vbll, vbll, vuc);
// CHECK: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>
// CHECK-LE: xor <16 x i8>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>

  res_vf = vec_round(vf);
// CHECK: call <4 x float> @llvm.round.v4f32(<4 x float>
// CHECK-LE: call <4 x float> @llvm.round.v4f32(<4 x float>

  res_vd = vec_round(vd);
// CHECK: call <2 x double> @llvm.round.v2f64(<2 x double>
// CHECK-LE: call <2 x double> @llvm.round.v2f64(<2 x double>

  res_vd = vec_perm(vd, vd, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vd = vec_splat(vd, 1);
// CHECK: [[T1:%.+]] = bitcast <2 x double> {{.+}} to <4 x i32>
// CHECK: [[T2:%.+]] = bitcast <2 x double> {{.+}} to <4 x i32>
// CHECK: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>
// CHECK-LE: xor <16 x i8>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x double> {{.+}} to <4 x i32>
// CHECK-LE: [[T2:%.+]] = bitcast <2 x double> {{.+}} to <4 x i32>
// CHECK-LE: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>

  res_vbll = vec_splat(vbll, 1);
// CHECK: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>
// CHECK-LE: xor <16 x i8>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>

  res_vsll =  vec_splat(vsll, 1);
// CHECK: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>
// CHECK-LE: xor <16 x i8>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>

  res_vull =  vec_splat(vull, 1);
// CHECK: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>
// CHECK-LE: xor <16 x i8>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: [[T2:%.+]] = bitcast <2 x i64> {{.+}} to <4 x i32>
// CHECK-LE: call <4 x i32> @llvm.ppc.altivec.vperm(<4 x i32> [[T1]], <4 x i32> [[T2]], <16 x i8>

  res_vsi = vec_pack(vsll, vsll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vui = vec_pack(vull, vull);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vbi = vec_pack(vbll, vbll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vsll = vec_vperm(vsll, vsll, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_vperm(vull, vull, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vd = vec_vperm(vd, vd, vuc);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  /* vec_vsx_ld */

  res_vsi = vec_vsx_ld(0, &vsi);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vui = vec_vsx_ld(0, &vui);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vf = vec_vsx_ld (0, &vf);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vsll = vec_vsx_ld(0, &vsll);
// CHECK: @llvm.ppc.vsx.lxvd2x
// CHECK-LE: @llvm.ppc.vsx.lxvd2x

  res_vull = vec_vsx_ld(0, &vull);
// CHECK: @llvm.ppc.vsx.lxvd2x
// CHECK-LE: @llvm.ppc.vsx.lxvd2x

  res_vd = vec_vsx_ld(0, &vd);
// CHECK: @llvm.ppc.vsx.lxvd2x
// CHECK-LE: @llvm.ppc.vsx.lxvd2x

  res_vull = vec_vsx_ld(0, &vull);
// CHECK: @llvm.ppc.vsx.lxvd2x
// CHECK-LE: @llvm.ppc.vsx.lxvd2x

  res_vd = vec_vsx_ld(0, &vd);
// CHECK: @llvm.ppc.vsx.lxvd2x
// CHECK-LE: @llvm.ppc.vsx.lxvd2x

  res_vss = vec_vsx_ld(0, &vss);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vss = vec_vsx_ld(0, &ss);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vus = vec_vsx_ld(0, &vus);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vus = vec_vsx_ld(0, &us);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vbc = vec_vsx_ld(0, &vbc);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vsc = vec_vsx_ld(0, &vsc);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vuc = vec_vsx_ld(0, &vuc);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vsc = vec_vsx_ld(0, &sc);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  res_vuc = vec_vsx_ld(0, &uc);
// CHECK: @llvm.ppc.vsx.lxvw4x
// CHECK-LE: @llvm.ppc.vsx.lxvw4x

  /* vec_vsx_st */

  vec_vsx_st(vsi, 0, &res_vsi);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vsi, 0, &res_si);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vui, 0, &res_vui);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vui, 0, &res_ui);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vf, 0, &res_vf);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vsll, 0, &res_vsll);
// CHECK: @llvm.ppc.vsx.stxvd2x
// CHECK-LE: @llvm.ppc.vsx.stxvd2x

  vec_vsx_st(vull, 0, &res_vull);
// CHECK: @llvm.ppc.vsx.stxvd2x
// CHECK-LE: @llvm.ppc.vsx.stxvd2x

  vec_vsx_st(vd, 0, &res_vd);
// CHECK: @llvm.ppc.vsx.stxvd2x
// CHECK-LE: @llvm.ppc.vsx.stxvd2x

  vec_vsx_st(vss, 0, &res_vss);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vss, 0, &res_ss);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vus, 0, &res_vus);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vus, 0, &res_us);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vsc, 0, &res_vsc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vsc, 0, &res_sc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vuc, 0, &res_vuc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vuc, 0, &res_uc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vbc, 0, &res_vbc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vbc, 0, &res_sc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vbc, 0, &res_uc);
// CHECK: @llvm.ppc.vsx.stxvw4x
// CHECK-LE: @llvm.ppc.vsx.stxvw4x

  /* vec_and */
  res_vsll = vec_and(vsll, vsll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_and(vbll, vsll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_and(vsll, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_and(vull, vull);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_and(vbll, vull);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_and(vull, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vbll = vec_and(vbll, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  /* vec_vand */
  res_vsll = vec_vand(vsll, vsll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_vand(vbll, vsll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_vand(vsll, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_vand(vull, vull);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_vand(vbll, vull);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_vand(vull, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vbll = vec_vand(vbll, vbll);
// CHECK: and <2 x i64>
// CHECK-LE: and <2 x i64>

  /* vec_andc */
  res_vsll = vec_andc(vsll, vsll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_andc(vbll, vsll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vsll = vec_andc(vsll, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_andc(vull, vull);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_andc(vbll, vull);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vull = vec_andc(vull, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vbll = vec_andc(vbll, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>
// CHECK-LE: xor <2 x i64>
// CHECK-LE: and <2 x i64>

  res_vf = vec_floor(vf);
// CHECK: call <4 x float> @llvm.floor.v4f32(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.floor.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_floor(vd);
// CHECK: call <2 x double> @llvm.floor.v2f64(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.floor.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_madd(vf, vf, vf);
// CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})

  res_vd = vec_madd(vd, vd, vd);
// CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})

  /* vec_mergeh */
  res_vsll = vec_mergeh(vsll, vsll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vsll = vec_mergeh(vsll, vbll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vsll = vec_mergeh(vbll, vsll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergeh(vull, vull);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergeh(vull, vbll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergeh(vbll, vull);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  /* vec_mergel */
  res_vsll = vec_mergel(vsll, vsll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vsll = vec_mergel(vsll, vbll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vsll = vec_mergel(vbll, vsll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergel(vull, vull);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergel(vull, vbll);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  res_vull = vec_mergel(vbll, vull);
// CHECK: @llvm.ppc.altivec.vperm
// CHECK-LE: @llvm.ppc.altivec.vperm

  /* vec_msub */
  res_vf = vec_msub(vf, vf, vf);
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>
// CHECK-LE: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-LE-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>

  res_vd = vec_msub(vd, vd, vd);
// CHECK: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>
// CHECK-LE: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-LE-NEXT: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>

  res_vsll = vec_mul(vsll, vsll);
// CHECK: mul <2 x i64>
// CHECK-LE: mul <2 x i64>

  res_vull = vec_mul(vull, vull);
// CHECK: mul <2 x i64>
// CHECK-LE: mul <2 x i64>

  res_vf = vec_mul(vf, vf);
// CHECK: fmul <4 x float> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: fmul <4 x float> %{{[0-9]+}}, %{{[0-9]+}}

  res_vd = vec_mul(vd, vd);
// CHECK: fmul <2 x double> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: fmul <2 x double> %{{[0-9]+}}, %{{[0-9]+}}

  res_vf = vec_nearbyint(vf);
// CHECK: call <4 x float> @llvm.round.v4f32(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.round.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_nearbyint(vd);
// CHECK: call <2 x double> @llvm.round.v2f64(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.round.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_nmadd(vf, vf, vf);
// CHECK: [[FM:[0-9]+]] = call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})
// CHECK-NEXT: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %[[FM]]
// CHECK-LE: [[FM:[0-9]+]] = call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})
// CHECK-LE-NEXT: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %[[FM]]

  res_vd = vec_nmadd(vd, vd, vd);
// CHECK: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})
// CHECK-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]
// CHECK-LE: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})
// CHECK-LE-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]

  res_vf = vec_nmsub(vf, vf, vf);
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-LE: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-LE-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>
// CHECK-LE: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}

  res_vd = vec_nmsub(vd, vd, vd);
// CHECK: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>
// CHECK-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]
// CHECK-LE: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-LE-NEXT: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>
// CHECK-LE-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]

  /* vec_nor */
  res_vsll = vec_nor(vsll, vsll);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>
// CHECK-LE: or <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_nor(vull, vull);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>
// CHECK-LE: or <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_nor(vbll, vbll);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>
// CHECK-LE: or <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vd = vec_nor(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: [[OR:%.+]] = or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-NEXT: xor <2 x i64> [[OR]], <i64 -1, i64 -1>
// CHECK-LE: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK-LE: [[OR:%.+]] = or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE-NEXT: xor <2 x i64> [[OR]], <i64 -1, i64 -1>

  /* vec_or */
  res_vsll = vec_or(vsll, vsll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vsll = vec_or(vbll, vsll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vsll = vec_or(vsll, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_or(vull, vull);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_or(vbll, vull);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_or(vull, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vbll = vec_or(vbll, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vd = vec_or(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK-LE: or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}

  res_vd = vec_or(vbll, vd);
// CHECK: [[T1:%.+]] = bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: [[T2:%.+]] = or <2 x i64> %{{[0-9]+}}, [[T1]]
// CHECK: bitcast <2 x i64> [[T2]] to <2 x double>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK-LE: [[T2:%.+]] = or <2 x i64> %{{[0-9]+}}, [[T1]]
// CHECK-LE: bitcast <2 x i64> [[T2]] to <2 x double>

  res_vd = vec_or(vd, vbll);
// CHECK: [[T1:%.+]] = bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: [[T2:%.+]] = or <2 x i64> [[T1]], %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[T2]] to <2 x double>
// CHECK-LE: [[T1:%.+]] = bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK-LE: [[T2:%.+]] = or <2 x i64> [[T1]], %{{[0-9]+}}
// CHECK-LE: bitcast <2 x i64> [[T2]] to <2 x double>

  res_vf = vec_re(vf);
// CHECK: call <4 x float> @llvm.ppc.vsx.xvresp(<4 x float>
// CHECK-LE: call <4 x float> @llvm.ppc.vsx.xvresp(<4 x float>

  res_vd = vec_re(vd);
// CHECK: call <2 x double> @llvm.ppc.vsx.xvredp(<2 x double>
// CHECK-LE: call <2 x double> @llvm.ppc.vsx.xvredp(<2 x double>

  res_vf = vec_rint(vf);
// CHECK: call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_rint(vd);
// CHECK: call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_rsqrte(vf);
// CHECK: call <4 x float> @llvm.ppc.vsx.xvrsqrtesp(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.ppc.vsx.xvrsqrtesp(<4 x float> %{{[0-9]+}})

  res_vd = vec_rsqrte(vd);
// CHECK: call <2 x double> @llvm.ppc.vsx.xvrsqrtedp(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.ppc.vsx.xvrsqrtedp(<2 x double> %{{[0-9]+}})

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vf = vec_sel(vd, vd, vbll);
// CHECK: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64> %{{[0-9]+}},
// CHECK: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: or <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>
// CHECK-LE: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK-LE: and <2 x i64> %{{[0-9]+}},
// CHECK-LE: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: or <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_sel(vd, vd, vull);
// CHECK: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64> %{{[0-9]+}},
// CHECK: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: or <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>
// CHECK-LE: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK-LE: and <2 x i64> %{{[0-9]+}},
// CHECK-LE: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: or <2 x i64>
// CHECK-LE: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>

  res_vf = vec_sqrt(vf);
// CHECK: call <4 x float> @llvm.sqrt.v4f32(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.sqrt.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_sqrt(vd);
// CHECK: call <2 x double> @llvm.sqrt.v2f64(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.sqrt.v2f64(<2 x double> %{{[0-9]+}})

  res_vd = vec_sub(vd, vd);
// CHECK: fsub <2 x double> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: fsub <2 x double> %{{[0-9]+}}, %{{[0-9]+}}

  res_vf = vec_trunc(vf);
// CHECK: call <4 x float> @llvm.trunc.v4f32(<4 x float> %{{[0-9]+}})
// CHECK-LE: call <4 x float> @llvm.trunc.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_trunc(vd);
// CHECK: call <2 x double> @llvm.trunc.v2f64(<2 x double> %{{[0-9]+}})
// CHECK-LE: call <2 x double> @llvm.trunc.v2f64(<2 x double> %{{[0-9]+}})

  /* vec_vor */
  res_vsll = vec_vor(vsll, vsll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vsll = vec_vor(vbll, vsll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vsll = vec_vor(vsll, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_vor(vull, vull);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_vor(vbll, vull);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vull = vec_vor(vull, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  res_vbll = vec_vor(vbll, vbll);
// CHECK: or <2 x i64>
// CHECK-LE: or <2 x i64>

  /* vec_xor */
  res_vsll = vec_xor(vsll, vsll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vsll = vec_xor(vbll, vsll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vsll = vec_xor(vsll, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_xor(vull, vull);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_xor(vbll, vull);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_xor(vull, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vbll = vec_xor(vbll, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_xor(vd, vd);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>
// CHECK-LE: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: bitcast <2 x i64> [[X1]] to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_xor(vd, vbll);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>
// CHECK-LE: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: bitcast <2 x i64> [[X1]] to <2 x double>

  dummy();
// CHECK: call void @dummy()
// CHECK-LE: call void @dummy()

  res_vd = vec_xor(vbll, vd);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>
// CHECK-LE: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-LE: bitcast <2 x i64> [[X1]] to <2 x double>

  /* vec_vxor */
  res_vsll = vec_vxor(vsll, vsll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vsll = vec_vxor(vbll, vsll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vsll = vec_vxor(vsll, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_vxor(vull, vull);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_vxor(vbll, vull);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vull = vec_vxor(vull, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vbll = vec_vxor(vbll, vbll);
// CHECK: xor <2 x i64>
// CHECK-LE: xor <2 x i64>

  res_vsll = vec_cts(vd, 0);
// CHECK: fmul <2 x double>
// CHECK: fptosi <2 x double> %{{.*}} to <2 x i64>
// CHECK-LE: fmul <2 x double>
// CHECK-LE: fptosi <2 x double> %{{.*}} to <2 x i64>

  res_vsll = vec_cts(vd, 31);
// CHECK: fmul <2 x double>
// CHECK: fptosi <2 x double> %{{.*}} to <2 x i64>
// CHECK-LE: fmul <2 x double>
// CHECK-LE: fptosi <2 x double> %{{.*}} to <2 x i64>

  res_vsll = vec_ctu(vd, 0);
// CHECK: fmul <2 x double>
// CHECK: fptoui <2 x double> %{{.*}} to <2 x i64>
// CHECK-LE: fmul <2 x double>
// CHECK-LE: fptoui <2 x double> %{{.*}} to <2 x i64>

  res_vsll = vec_ctu(vd, 31);
// CHECK: fmul <2 x double>
// CHECK: fptoui <2 x double> %{{.*}} to <2 x i64>
// CHECK-LE: fmul <2 x double>
// CHECK-LE: fptoui <2 x double> %{{.*}} to <2 x i64>

  res_vd = vec_ctf(vsll, 0);
// CHECK: sitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK: fmul <2 x double>
// CHECK-LE: sitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK-LE: fmul <2 x double>

  res_vd = vec_ctf(vsll, 31);
// CHECK: sitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK: fmul <2 x double>
// CHECK-LE: sitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK-LE: fmul <2 x double>

  res_vd = vec_ctf(vull, 0);
// CHECK: uitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK: fmul <2 x double>
// CHECK-LE: uitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK-LE: fmul <2 x double>

  res_vd = vec_ctf(vull, 31);
// CHECK: uitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK: fmul <2 x double>
// CHECK-LE: uitofp <2 x i64> %{{.*}} to <2 x double>
// CHECK-LE: fmul <2 x double>
}
Exemplo n.º 6
0
void test1() {
// CHECK-LABEL: define void @test1

  res_vd = vec_add(vd, vd);
// CHECK: fadd <2 x double>

  res_vd = vec_and(vbll, vd);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  res_vd = vec_and(vd, vbll);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  res_vd = vec_and(vd, vd);
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_andc(vbll, vd);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_andc(vd, vbll);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_andc(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]*}} to <2 x i64>
// CHECK: xor <2 x i64> %{{[0-9]*}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]*}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_ceil(vd);
// CHECK: call <2 x double> @llvm.ceil.v2f64(<2 x double> %{{[0-9]*}})

  res_vf = vec_ceil(vf);
// CHECK: call <4 x float> @llvm.ceil.v4f32(<4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpeq(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpeqdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpeq(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpeqsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpge(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpge(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmpgt(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmpgt(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmple(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgedp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmple(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgesp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  res_vbll = vec_cmplt(vd, vd);
// CHECK: call <2 x i64> @llvm.ppc.vsx.xvcmpgtdp(<2 x double> %{{[0-9]*}}, <2 x double> %{{[0-9]*}})

  res_vbi = vec_cmplt(vf, vf);
// CHECK: call <4 x i32> @llvm.ppc.vsx.xvcmpgtsp(<4 x float> %{{[0-9]*}}, <4 x float> %{{[0-9]*}})

  /* vec_div */
  res_vf = vec_div(vf, vf);
// CHECK: @llvm.ppc.vsx.xvdivsp

  res_vd = vec_div(vd, vd);
// CHECK: @llvm.ppc.vsx.xvdivdp

  /* vec_max */
  res_vf = vec_max(vf, vf);
// CHECK: @llvm.ppc.vsx.xvmaxsp

  res_vd = vec_max(vd, vd);
// CHECK: @llvm.ppc.vsx.xvmaxdp

  res_vf = vec_vmaxfp(vf, vf);
// CHECK: @llvm.ppc.vsx.xvmaxsp

  /* vec_min */
  res_vf = vec_min(vf, vf);
// CHECK: @llvm.ppc.vsx.xvminsp

  res_vd = vec_min(vd, vd);
// CHECK: @llvm.ppc.vsx.xvmindp

  res_vf = vec_vminfp(vf, vf);
// CHECK: @llvm.ppc.vsx.xvminsp

  res_d = __builtin_vsx_xsmaxdp(d, d);
// CHECK: @llvm.ppc.vsx.xsmaxdp

  res_d = __builtin_vsx_xsmindp(d, d);
// CHECK: @llvm.ppc.vsx.xsmindp

  /* vec_perm */
  res_vsll = vec_perm(vsll, vsll, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  res_vull = vec_perm(vull, vull, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  res_vd = vec_perm(vd, vd, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  res_vsll = vec_vperm(vsll, vsll, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  res_vull = vec_vperm(vull, vull, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  res_vd = vec_vperm(vd, vd, vuc);
// CHECK: @llvm.ppc.altivec.vperm

  /* vec_vsx_ld */

  res_vsi = vec_vsx_ld(0, &vsi);
// CHECK: @llvm.ppc.vsx.lxvw4x

  res_vui = vec_vsx_ld(0, &vui);
// CHECK: @llvm.ppc.vsx.lxvw4x

  res_vf = vec_vsx_ld (0, &vf);
// CHECK: @llvm.ppc.vsx.lxvw4x

  res_vsll = vec_vsx_ld(0, &vsll);
// CHECK: @llvm.ppc.vsx.lxvd2x

  res_vull = vec_vsx_ld(0, &vull);
// CHECK: @llvm.ppc.vsx.lxvd2x

  res_vd = vec_vsx_ld(0, &vd);
// CHECK: @llvm.ppc.vsx.lxvd2x

  /* vec_vsx_st */

  vec_vsx_st(vsi, 0, &res_vsi);
// CHECK: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vui, 0, &res_vui);
// CHECK: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vf, 0, &res_vf);
// CHECK: @llvm.ppc.vsx.stxvw4x

  vec_vsx_st(vsll, 0, &res_vsll);
// CHECK: @llvm.ppc.vsx.stxvd2x

  vec_vsx_st(vull, 0, &res_vull);
// CHECK: @llvm.ppc.vsx.stxvd2x

  vec_vsx_st(vd, 0, &res_vd);
// CHECK: @llvm.ppc.vsx.stxvd2x

  /* vec_and */
  res_vsll = vec_and(vsll, vsll);
// CHECK: and <2 x i64>

  res_vsll = vec_and(vbll, vsll);
// CHECK: and <2 x i64>

  res_vsll = vec_and(vsll, vbll);
// CHECK: and <2 x i64>

  res_vull = vec_and(vull, vull);
// CHECK: and <2 x i64>

  res_vull = vec_and(vbll, vull);
// CHECK: and <2 x i64>

  res_vull = vec_and(vull, vbll);
// CHECK: and <2 x i64>

  res_vbll = vec_and(vbll, vbll);
// CHECK: and <2 x i64>

  /* vec_vand */
  res_vsll = vec_vand(vsll, vsll);
// CHECK: and <2 x i64>

  res_vsll = vec_vand(vbll, vsll);
// CHECK: and <2 x i64>

  res_vsll = vec_vand(vsll, vbll);
// CHECK: and <2 x i64>

  res_vull = vec_vand(vull, vull);
// CHECK: and <2 x i64>

  res_vull = vec_vand(vbll, vull);
// CHECK: and <2 x i64>

  res_vull = vec_vand(vull, vbll);
// CHECK: and <2 x i64>

  res_vbll = vec_vand(vbll, vbll);
// CHECK: and <2 x i64>

  /* vec_andc */
  res_vsll = vec_andc(vsll, vsll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vsll = vec_andc(vbll, vsll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vsll = vec_andc(vsll, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vull = vec_andc(vull, vull);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vull = vec_andc(vbll, vull);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vull = vec_andc(vull, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vbll = vec_andc(vbll, vbll);
// CHECK: xor <2 x i64>
// CHECK: and <2 x i64>

  res_vf = vec_floor(vf);
// CHECK: call <4 x float> @llvm.floor.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_floor(vd);
// CHECK: call <2 x double> @llvm.floor.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_madd(vf, vf, vf);
// CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})

  res_vd = vec_madd(vd, vd, vd);
// CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})

  res_vf = vec_msub(vf, vf, vf);
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>

  res_vd = vec_msub(vd, vd, vd);
// CHECK: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>

  res_vf = vec_mul(vf, vf);
// CHECK: fmul <4 x float> %{{[0-9]+}}, %{{[0-9]+}}

  res_vd = vec_mul(vd, vd);
// CHECK: fmul <2 x double> %{{[0-9]+}}, %{{[0-9]+}}

  res_vf = vec_nearbyint(vf);
// CHECK: call <4 x float> @llvm.round.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_nearbyint(vd);
// CHECK: call <2 x double> @llvm.round.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_nmadd(vf, vf, vf);
// CHECK: [[FM:[0-9]+]] = call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}})
// CHECK-NEXT: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %[[FM]]

  res_vd = vec_nmadd(vd, vd, vd);
// CHECK: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}})
// CHECK-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]

  res_vf = vec_nmsub(vf, vf, vf);
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{[0-9]+}}, <4 x float> %{{[0-9]+}}, <4 x float>
// CHECK: fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %{{[0-9]+}}

  res_vd = vec_nmsub(vd, vd, vd);
// CHECK: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %{{[0-9]+}}
// CHECK-NEXT: [[FM:[0-9]+]] = call <2 x double> @llvm.fma.v2f64(<2 x double> %{{[0-9]+}}, <2 x double> %{{[0-9]+}}, <2 x double>
// CHECK-NEXT: fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %[[FM]]

  /* vec_nor */
  res_vsll = vec_nor(vsll, vsll);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>

  res_vull = vec_nor(vull, vull);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>

  res_vull = vec_nor(vbll, vbll);
// CHECK: or <2 x i64>
// CHECK: xor <2 x i64>

  res_vd = vec_nor(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: [[OR:%.+]] = or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK-NEXT: xor <2 x i64> [[OR]], <i64 -1, i64 -1>

  /* vec_or */
  res_vsll = vec_or(vsll, vsll);
// CHECK: or <2 x i64>

  res_vsll = vec_or(vbll, vsll);
// CHECK: or <2 x i64>

  res_vsll = vec_or(vsll, vbll);
// CHECK: or <2 x i64>

  res_vull = vec_or(vull, vull);
// CHECK: or <2 x i64>

  res_vull = vec_or(vbll, vull);
// CHECK: or <2 x i64>

  res_vull = vec_or(vull, vbll);
// CHECK: or <2 x i64>

  res_vbll = vec_or(vbll, vbll);
// CHECK: or <2 x i64>

  res_vd = vec_or(vd, vd);
// CHECK: bitcast <2 x double> %{{[0-9]+}} to <2 x i64>
// CHECK: or <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}

  res_vf = vec_rint(vf);
// CHECK: call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_rint(vd);
// CHECK: call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %{{[0-9]+}})

  res_vf = vec_rsqrte(vf);
// CHECK: call <4 x float> @llvm.ppc.vsx.xvrsqrtesp(<4 x float> %{{[0-9]+}})

  res_vd = vec_rsqrte(vd);
// CHECK: call <2 x double> @llvm.ppc.vsx.xvrsqrtedp(<2 x double> %{{[0-9]+}})

  dummy();
// CHECK: call void @dummy()

  res_vf = vec_sel(vd, vd, vbll);
// CHECK: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64> %{{[0-9]+}},
// CHECK: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: or <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_sel(vd, vd, vull);
// CHECK: xor <2 x i64> %{{[0-9]+}}, <i64 -1, i64 -1>
// CHECK: and <2 x i64> %{{[0-9]+}},
// CHECK: and <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: or <2 x i64>
// CHECK: bitcast <2 x i64> %{{[0-9]+}} to <2 x double>

  res_vf = vec_sqrt(vf);
// CHECK: call <4 x float> @llvm.sqrt.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_sqrt(vd);
// CHECK: call <2 x double> @llvm.sqrt.v2f64(<2 x double> %{{[0-9]+}})

  res_vd = vec_sub(vd, vd);
// CHECK: fsub <2 x double> %{{[0-9]+}}, %{{[0-9]+}}

  res_vf = vec_trunc(vf);
// CHECK: call <4 x float> @llvm.trunc.v4f32(<4 x float> %{{[0-9]+}})

  res_vd = vec_trunc(vd);
// CHECK: call <2 x double> @llvm.trunc.v2f64(<2 x double> %{{[0-9]+}})

  /* vec_vor */
  res_vsll = vec_vor(vsll, vsll);
// CHECK: or <2 x i64>

  res_vsll = vec_vor(vbll, vsll);
// CHECK: or <2 x i64>

  res_vsll = vec_vor(vsll, vbll);
// CHECK: or <2 x i64>

  res_vull = vec_vor(vull, vull);
// CHECK: or <2 x i64>

  res_vull = vec_vor(vbll, vull);
// CHECK: or <2 x i64>

  res_vull = vec_vor(vull, vbll);
// CHECK: or <2 x i64>

  res_vbll = vec_vor(vbll, vbll);
// CHECK: or <2 x i64>

  /* vec_xor */
  res_vsll = vec_xor(vsll, vsll);
// CHECK: xor <2 x i64>

  res_vsll = vec_xor(vbll, vsll);
// CHECK: xor <2 x i64>

  res_vsll = vec_xor(vsll, vbll);
// CHECK: xor <2 x i64>

  res_vull = vec_xor(vull, vull);
// CHECK: xor <2 x i64>

  res_vull = vec_xor(vbll, vull);
// CHECK: xor <2 x i64>

  res_vull = vec_xor(vull, vbll);
// CHECK: xor <2 x i64>

  res_vbll = vec_xor(vbll, vbll);
// CHECK: xor <2 x i64>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_xor(vd, vd);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_xor(vd, vbll);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>

  dummy();
// CHECK: call void @dummy()

  res_vd = vec_xor(vbll, vd);
// CHECK: [[X1:%.+]] = xor <2 x i64> %{{[0-9]+}}, %{{[0-9]+}}
// CHECK: bitcast <2 x i64> [[X1]] to <2 x double>

  /* vec_vxor */
  res_vsll = vec_vxor(vsll, vsll);
// CHECK: xor <2 x i64>

  res_vsll = vec_vxor(vbll, vsll);
// CHECK: xor <2 x i64>

  res_vsll = vec_vxor(vsll, vbll);
// CHECK: xor <2 x i64>

  res_vull = vec_vxor(vull, vull);
// CHECK: xor <2 x i64>

  res_vull = vec_vxor(vbll, vull);
// CHECK: xor <2 x i64>

  res_vull = vec_vxor(vull, vbll);
// CHECK: xor <2 x i64>

  res_vbll = vec_vxor(vbll, vbll);
// CHECK: xor <2 x i64>

}
Exemplo n.º 7
0
vector double
test2 (vector double x, vector double y)
{
  return vec_div (x, y);
}
Exemplo n.º 8
0
/* a vectorized version of the Voigt function using Altivec / VMX instructions */
void my_voigt(const float *damping, const float *frequency_offset, float *voigt_value, int N)
{
   // coefficients of the rational approximation formula
   // to the complementary error function
   const vector float A0 = (vector float) (122.607931777104326f);
   const vector float A1 = (vector float) (214.382388694706425f);
   const vector float A2 = (vector float) (181.928533092181549f);
   const vector float A3 = (vector float) (93.155580458138441f);
   const vector float A4 = (vector float) (30.180142196210589f);
   const vector float A5 = (vector float) (5.912626209773153f);
   const vector float A6 = (vector float) (0.564189583562615f);
   const vector float B0 = (vector float) (122.60793177387535f);
   const vector float B1 = (vector float) (352.730625110963558f);
   const vector float B2 = (vector float) (457.334478783897737f);
   const vector float B3 = (vector float) (348.703917719495792f);
   const vector float B4 = (vector float) (170.354001821091472f);
   const vector float B5 = (vector float) (53.992906912940207f);
   const vector float B6 = (vector float) (10.479857114260399f);

   vector float ivsigno;
   vector float V;
   vector float Z1_real;
   vector float Z1_imag;
   vector float Z2_real;
   vector float Z2_imag;
   vector float Z3_real;
   vector float Z3_imag;
   vector float Z4_real;
   vector float Z4_imag;
   vector float Z5_real;
   vector float Z5_imag;
   vector float Z6_real;
   vector float Z6_imag;
   vector float ZZ1_real;
   vector float ZZ1_imag;
   vector float ZZ2_real;
   vector float ZZ2_imag;
   vector float ZZ3_real;
   vector float ZZ3_imag;
   vector float ZZ4_real;
   vector float ZZ4_imag;
   vector float ZZ5_real;
   vector float ZZ5_imag;
   vector float ZZ6_real;
   vector float ZZ6_imag;
   vector float ZZ7_real;
   vector float ZZ7_imag;
   vector float division_factor;
   vector float ZZZ_real;

   vector bool int mask;
   const vector float one = (vector float) (1.0f);
   const vector float zero = (vector float) (-0.0f);
   const vector float mone = (vector float) (-1.0f);

   vector float damp;
   vector float offs;

   for(int i=0; i<N; i+=4){
      damp = vec_ld(0,(float *) &damping[i]);
      offs = vec_ld(0,(float *) &frequency_offset[i]);
      mask = vec_cmplt(offs,zero);
      ivsigno = vec_sel(mone, one, mask);
      //ivsigno = (vector float) (1.0f);
      V = vec_madd(ivsigno, offs, zero);
      Z1_real = vec_madd(A6, damp, A5);
      Z1_imag = vec_nmsub(A6, V, zero);
      Z2_real = vec_add(vec_madd(Z1_real,damp,zero),vec_madd(Z1_imag,V,A4)); 
      Z2_imag = vec_add(vec_nmsub(Z1_real,V,zero),vec_madd(Z1_imag,damp,zero));
      Z3_real = vec_add(vec_madd(Z2_real,damp,zero),vec_madd(Z2_imag,V,A3));
      Z3_imag = vec_add(vec_nmsub(Z2_real,V,zero),vec_madd(Z2_imag,damp,zero));
      Z4_real = vec_add(vec_madd(Z3_real,damp,zero),vec_madd(Z3_imag,V,A2));
      Z4_imag = vec_add(vec_nmsub(Z3_real,V,zero),vec_madd(Z3_imag,damp,zero));
      Z5_real = vec_add(vec_madd(Z4_real,damp,zero),vec_madd(Z4_imag,V,A1));
      Z5_imag = vec_add(vec_nmsub(Z4_real,V,zero),vec_madd(Z4_imag,damp,zero));
      Z6_real = vec_add(vec_madd(Z5_real,damp,zero),vec_madd(Z5_imag,V,A0));
      Z6_imag = vec_add(vec_nmsub(Z5_real,V,zero),vec_madd(Z5_imag,damp,zero));
      ZZ1_real = vec_add(damp,B6); 
      ZZ1_imag = vec_madd(mone,V,zero);
      ZZ2_real = vec_add(vec_madd(ZZ1_real,damp,zero),vec_madd(ZZ1_imag,V,B5));
      ZZ2_imag = vec_add(vec_nmsub(ZZ1_real,V,zero),vec_madd(ZZ1_imag,damp,zero));
      ZZ3_real = vec_add(vec_madd(ZZ2_real,damp,zero),vec_madd(ZZ2_imag,V,B4));
      ZZ3_imag = vec_add(vec_nmsub(ZZ2_real,V,zero),vec_madd(ZZ2_imag,damp,zero));
      ZZ4_real = vec_add(vec_madd(ZZ3_real,damp,zero),vec_madd(ZZ3_imag,V,B3));
      ZZ4_imag = vec_add(vec_nmsub(ZZ3_real,V,zero),vec_madd(ZZ3_imag,damp,zero));
      ZZ5_real = vec_add(vec_madd(ZZ4_real,damp,zero),vec_madd(ZZ4_imag,V,B2));
      ZZ5_imag = vec_add(vec_nmsub(ZZ4_real,V,zero),vec_madd(ZZ4_imag,damp,zero));
      ZZ6_real = vec_add(vec_madd(ZZ5_real,damp,zero),vec_madd(ZZ5_imag,V,B1));
      ZZ6_imag = vec_add(vec_nmsub(ZZ5_real,V,zero),vec_madd(ZZ5_imag,damp,zero));
      ZZ7_real = vec_add(vec_madd(ZZ6_real,damp,zero),vec_madd(ZZ6_imag,V,B0));
      ZZ7_imag = vec_add(vec_nmsub(ZZ6_real,V,zero),vec_madd(ZZ6_imag,damp,zero));
      division_factor = vec_div(one,vec_madd(ZZ7_real,ZZ7_real,vec_madd(ZZ7_imag,ZZ7_imag,zero)));
      ZZZ_real = vec_madd(vec_madd(Z6_real,ZZ7_real,vec_madd(Z6_imag,ZZ7_imag,zero)),division_factor,zero); 
      vec_st(ZZZ_real,0,(float *)&voigt_value[i]);
   }
}
Exemplo n.º 9
0
 VSX_FINLINE(vec_double2) vec_rsqrt(const vec_double2& a)
 { return vec_div(vec_double2_sp(1), vec_sqrt(a)); }
Exemplo n.º 10
0
 VSX_FINLINE(vec_float4) vec_rsqrt(const vec_float4& a)
 { return vec_div(vec_float4_sp(1), vec_sqrt(a)); }