C++ (Cpp) vsetq_lane_u32 Examples

Example #1

File: vsetQ_laneu32.c Project: Akheon23/chromecast-mirrored-source.toolchain

void test_vsetQ_laneu32 (void)
{
  uint32x4_t out_uint32x4_t;
  uint32_t arg0_uint32_t;
  uint32x4_t arg1_uint32x4_t;

  out_uint32x4_t = vsetq_lane_u32 (arg0_uint32_t, arg1_uint32x4_t, 1);
}

Example #2

File: aarch64-neon-copy.c Project: Xmister/clang-onex

uint32x4_t test_vsetq_lane_u32(uint32_t v1, uint32x4_t v2) {
   // CHECK: test_vsetq_lane_u32
  return vsetq_lane_u32(v1, v2, 1);
  // CHECK: ins {{v[0-9]+}}.s[1], {{w[0-9]+}}
}

Example #3

File: aarch64-neon-vget.c Project: Blizzard/clang

uint32x4_t test_vsetq_lane_u32(uint32_t a, uint32x4_t b) {
  // CHECK-LABEL: test_vsetq_lane_u32:
  // CHECK-NEXT:  ins.s v0[3], w0
  // CHECK-NEXT:  ret
  return vsetq_lane_u32(a, b, 3);
}

Example #4

File: neon-simd.cpp Project: Mellnik/hash-plugin

bool CPU_ProbeNEON()
{
#if defined(CRYPTOPP_NO_CPU_FEATURE_PROBES)
	return false;
#elif (CRYPTOPP_ARM_NEON_AVAILABLE)
# if defined(CRYPTOPP_MS_STYLE_INLINE_ASSEMBLY)
	volatile bool result = true;
	__try
	{
		uint32_t v1[4] = {1,1,1,1};
		uint32x4_t x1 = vld1q_u32(v1);
		uint64_t v2[2] = {1,1};
		uint64x2_t x2 = vld1q_u64(v2);

		uint32x4_t x3 = vdupq_n_u32(2);
		x3 = vsetq_lane_u32(vgetq_lane_u32(x1,0),x3,0);
		x3 = vsetq_lane_u32(vgetq_lane_u32(x1,3),x3,3);
		uint64x2_t x4 = vdupq_n_u64(2);
		x4 = vsetq_lane_u64(vgetq_lane_u64(x2,0),x4,0);
		x4 = vsetq_lane_u64(vgetq_lane_u64(x2,1),x4,1);

		result = !!(vgetq_lane_u32(x3,0) | vgetq_lane_u64(x4,1));
	}
	__except (EXCEPTION_EXECUTE_HANDLER)
	{
		return false;
	}
	return result;
# else

	// longjmp and clobber warnings. Volatile is required.
	// http://github.com/weidai11/cryptopp/issues/24 and http://stackoverflow.com/q/7721854
	volatile bool result = true;

	volatile SigHandler oldHandler = signal(SIGILL, SigIllHandler);
	if (oldHandler == SIG_ERR)
		return false;

	volatile sigset_t oldMask;
	if (sigprocmask(0, NULLPTR, (sigset_t*)&oldMask))
		return false;

	if (setjmp(s_jmpSIGILL))
		result = false;
	else
	{
		uint32_t v1[4] = {1,1,1,1};
		uint32x4_t x1 = vld1q_u32(v1);
		uint64_t v2[2] = {1,1};
		uint64x2_t x2 = vld1q_u64(v2);

		uint32x4_t x3 = {0,0,0,0};
		x3 = vsetq_lane_u32(vgetq_lane_u32(x1,0),x3,0);
		x3 = vsetq_lane_u32(vgetq_lane_u32(x1,3),x3,3);
		uint64x2_t x4 = {0,0};
		x4 = vsetq_lane_u64(vgetq_lane_u64(x2,0),x4,0);
		x4 = vsetq_lane_u64(vgetq_lane_u64(x2,1),x4,1);

		// Hack... GCC optimizes away the code and returns true
		result = !!(vgetq_lane_u32(x3,0) | vgetq_lane_u64(x4,1));
	}

	sigprocmask(SIG_SETMASK, (sigset_t*)&oldMask, NULLPTR);
	signal(SIGILL, oldHandler);
	return result;
# endif
#else
	return false;
#endif  // CRYPTOPP_ARM_NEON_AVAILABLE
}

Example #5

File: aarch64-neon-vget.c Project: AlexDenisov/clang

// CHECK-LABEL: define <4 x i32> @test_vsetq_lane_u32(i32 %a, <4 x i32> %b) #0 {
// CHECK:   [[TMP0:%.*]] = bitcast <4 x i32> %b to <16 x i8>
// CHECK:   [[TMP1:%.*]] = bitcast <16 x i8> [[TMP0]] to <4 x i32>
// CHECK:   [[VSET_LANE:%.*]] = insertelement <4 x i32> [[TMP1]], i32 %a, i32 3
// CHECK:   ret <4 x i32> [[VSET_LANE]]
uint32x4_t test_vsetq_lane_u32(uint32_t a, uint32x4_t b) {
  return vsetq_lane_u32(a, b, 3);
}